Eliminate G8R8 conversion, native is actualy R8G8

morton: Fixup compiler warning

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 "${CLANG_FORMAT} -i @(Get-ChildItem -Recurse ${SRCS}/* -Include \'*.h\', \'*.cpp\')"
+            COMMAND powershell.exe -Command "Get-ChildItem ${SRCS}/* -Include *.cpp,*.h -Recurse | Foreach {${CLANG_FORMAT} -i $_.fullname}"
             COMMENT ${CCOMMENT})
     elseif(MINGW)
         add_custom_target(clang-format

src/core/file_sys/patch_manager.cpp

@@ -26,6 +26,11 @@ 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);
@@ -57,6 +62,15 @@ 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
@@ -70,6 +84,30 @@ 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;
 }
 
@@ -314,18 +352,25 @@ 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

@@ -71,10 +71,6 @@ 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
@@ -260,6 +256,36 @@ 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 = static_cast<u32>(val[0]);
+    }
+}
+
 /**
  * Turns hex string character into the equivalent byte.
  *
@@ -409,6 +435,27 @@ 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;
@@ -599,8 +646,7 @@ 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 += ",";
+                val += fmt::format("{:x},", thread->GetThreadID());
             }
         }
         val.pop_back();
@@ -791,11 +837,15 @@ 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) {
-        LongToGdbHex(reply, RegRead(id, current_thread));
+        u128 r = FpuRead(id, current_thread);
+        LongToGdbHex(reply, r[0]);
+        LongToGdbHex(reply + 16, r[1]);
     } else if (id == FPCR_REGISTER) {
-        LongToGdbHex(reply, RegRead(TODO_DUMMY_REG_998, current_thread));
-    } else {
-        LongToGdbHex(reply, RegRead(TODO_DUMMY_REG_997, current_thread));
+        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));
     }
 
     SendReply(reinterpret_cast<char*>(reply));
@@ -822,13 +872,18 @@ static void ReadRegisters() {
 
     bufptr += 8;
 
-    for (u32 reg = UC_ARM64_REG_Q0; reg <= UC_ARM64_REG_Q0 + 31; reg++) {
-        LongToGdbHex(bufptr + reg * 16, RegRead(reg, current_thread));
+    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]);
     }
 
     bufptr += 32 * 32;
 
-    LongToGdbHex(bufptr, RegRead(TODO_DUMMY_REG_998, current_thread));
+    r = FpuRead(FPCR_REGISTER, current_thread);
+    IntToGdbHex(bufptr, static_cast<u32>(r[0]));
 
     bufptr += 8;
 
@@ -853,14 +908,12 @@ 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) {
-        RegWrite(id, GdbHexToLong(buffer_ptr), current_thread);
+        FpuWrite(id, GdbHexToU128(buffer_ptr), current_thread);
     } else if (id == FPCR_REGISTER) {
-        RegWrite(TODO_DUMMY_REG_998, GdbHexToLong(buffer_ptr), current_thread);
-    } else {
-        RegWrite(TODO_DUMMY_REG_997, GdbHexToLong(buffer_ptr), current_thread);
+    } else if (id == FPCR_REGISTER + 1) {
     }
 
-    // Update Unicorn context skipping scheduler, no running threads at this point
+    // Update ARM context, skipping scheduler - no running threads at this point
     Core::System::GetInstance()
         .ArmInterface(current_core)
         .LoadContext(current_thread->GetContext());
@@ -885,13 +938,13 @@ 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(TODO_DUMMY_REG_998, GdbHexToLong(buffer_ptr + i * 16), current_thread);
-        } else {
-            UNIMPLEMENTED();
+            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);
         }
     }
 
-    // Update Unicorn context skipping scheduler, no running threads at this point
+    // Update ARM context, skipping scheduler - no running threads at this point
     Core::System::GetInstance()
         .ArmInterface(current_core)
         .LoadContext(current_thread->GetContext());
@@ -917,12 +970,6 @@ 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");
     }
@@ -967,7 +1014,7 @@ void Break(bool is_memory_break) {
 static void Step() {
     if (command_length > 1) {
         RegWrite(PC_REGISTER, GdbHexToLong(command_buffer + 1), current_thread);
-        // Update Unicorn context skipping scheduler, no running threads at this point
+        // Update ARM context, skipping scheduler - no running threads at this point
         Core::System::GetInstance()
             .ArmInterface(current_core)
             .LoadContext(current_thread->GetContext());
@@ -1010,7 +1057,7 @@ 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});
@@ -1321,13 +1368,15 @@ void SetCpuStepFlag(bool is_step) {
 }
 
 void SendTrap(Kernel::Thread* thread, int trap) {
-    if (send_trap) {
-        if (!halt_loop || current_thread == thread) {
-            current_thread = thread;
-            SendSignal(thread, trap);
-        }
-        halt_loop = true;
-        send_trap = false;
+    if (!send_trap) {
+        return;
     }
+
+    if (!halt_loop || current_thread == thread) {
+        current_thread = thread;
+        SendSignal(thread, trap);
+    }
+    halt_loop = true;
+    send_trap = false;
 }
 }; // namespace GDBStub

src/core/hle/kernel/resource_limit.h

@@ -14,7 +14,7 @@ namespace Kernel {
 
 class KernelCore;
 
-enum class ResourceType {
+enum class ResourceType : u32 {
     PhysicalMemory,
     Threads,
     Events,
@@ -25,6 +25,10 @@ enum class ResourceType {
     ResourceTypeCount
 };
 
+constexpr bool IsValidResourceType(ResourceType type) {
+    return type < ResourceType::ResourceTypeCount;
+}
+
 class ResourceLimit final : public Object {
 public:
     /**

src/core/hle/kernel/svc.cpp

@@ -105,6 +105,38 @@ ResultCode MapUnmapMemorySanityChecks(const VMManager& vm_manager, VAddr dst_add
 
     return RESULT_SUCCESS;
 }
+
+enum class ResourceLimitValueType {
+    CurrentValue,
+    LimitValue,
+};
+
+ResultVal<s64> RetrieveResourceLimitValue(Handle resource_limit, u32 resource_type,
+                                          ResourceLimitValueType value_type) {
+    const auto type = static_cast<ResourceType>(resource_type);
+    if (!IsValidResourceType(type)) {
+        LOG_ERROR(Kernel_SVC, "Invalid resource limit type: '{}'", resource_type);
+        return ERR_INVALID_ENUM_VALUE;
+    }
+
+    const auto& kernel = Core::System::GetInstance().Kernel();
+    const auto* const current_process = kernel.CurrentProcess();
+    ASSERT(current_process != nullptr);
+
+    const auto resource_limit_object =
+        current_process->GetHandleTable().Get<ResourceLimit>(resource_limit);
+    if (!resource_limit_object) {
+        LOG_ERROR(Kernel_SVC, "Handle to non-existent resource limit instance used. Handle={:08X}",
+                  resource_limit);
+        return ERR_INVALID_HANDLE;
+    }
+
+    if (value_type == ResourceLimitValueType::CurrentValue) {
+        return MakeResult(resource_limit_object->GetCurrentResourceValue(type));
+    }
+
+    return MakeResult(resource_limit_object->GetMaxResourceValue(type));
+}
 } // Anonymous namespace
 
 /// Set the process heap to a given Size. It can both extend and shrink the heap.
@@ -671,7 +703,8 @@ static ResultCode GetInfo(u64* result, u64 info_id, u64 handle, u64 info_sub_id)
         break;
     }
     default:
-        UNIMPLEMENTED();
+        LOG_WARNING(Kernel_SVC, "(STUBBED) Unimplemented svcGetInfo id=0x{:016X}", info_id);
+        return ERR_INVALID_ENUM_VALUE;
     }
 
     return RESULT_SUCCESS;
@@ -1345,6 +1378,87 @@ static ResultCode GetProcessInfo(u64* out, Handle process_handle, u32 type) {
     return RESULT_SUCCESS;
 }
 
+static ResultCode CreateResourceLimit(Handle* out_handle) {
+    LOG_DEBUG(Kernel_SVC, "called");
+
+    auto& kernel = Core::System::GetInstance().Kernel();
+    auto resource_limit = ResourceLimit::Create(kernel);
+
+    auto* const current_process = kernel.CurrentProcess();
+    ASSERT(current_process != nullptr);
+
+    const auto handle = current_process->GetHandleTable().Create(std::move(resource_limit));
+    if (handle.Failed()) {
+        return handle.Code();
+    }
+
+    *out_handle = *handle;
+    return RESULT_SUCCESS;
+}
+
+static ResultCode GetResourceLimitLimitValue(u64* out_value, Handle resource_limit,
+                                             u32 resource_type) {
+    LOG_DEBUG(Kernel_SVC, "called. Handle={:08X}, Resource type={}", resource_limit, resource_type);
+
+    const auto limit_value = RetrieveResourceLimitValue(resource_limit, resource_type,
+                                                        ResourceLimitValueType::LimitValue);
+    if (limit_value.Failed()) {
+        return limit_value.Code();
+    }
+
+    *out_value = static_cast<u64>(*limit_value);
+    return RESULT_SUCCESS;
+}
+
+static ResultCode GetResourceLimitCurrentValue(u64* out_value, Handle resource_limit,
+                                               u32 resource_type) {
+    LOG_DEBUG(Kernel_SVC, "called. Handle={:08X}, Resource type={}", resource_limit, resource_type);
+
+    const auto current_value = RetrieveResourceLimitValue(resource_limit, resource_type,
+                                                          ResourceLimitValueType::CurrentValue);
+    if (current_value.Failed()) {
+        return current_value.Code();
+    }
+
+    *out_value = static_cast<u64>(*current_value);
+    return RESULT_SUCCESS;
+}
+
+static ResultCode SetResourceLimitLimitValue(Handle resource_limit, u32 resource_type, u64 value) {
+    LOG_DEBUG(Kernel_SVC, "called. Handle={:08X}, Resource type={}, Value={}", resource_limit,
+              resource_type, value);
+
+    const auto type = static_cast<ResourceType>(resource_type);
+    if (!IsValidResourceType(type)) {
+        LOG_ERROR(Kernel_SVC, "Invalid resource limit type: '{}'", resource_type);
+        return ERR_INVALID_ENUM_VALUE;
+    }
+
+    auto& kernel = Core::System::GetInstance().Kernel();
+    auto* const current_process = kernel.CurrentProcess();
+    ASSERT(current_process != nullptr);
+
+    auto resource_limit_object =
+        current_process->GetHandleTable().Get<ResourceLimit>(resource_limit);
+    if (!resource_limit_object) {
+        LOG_ERROR(Kernel_SVC, "Handle to non-existent resource limit instance used. Handle={:08X}",
+                  resource_limit);
+        return ERR_INVALID_HANDLE;
+    }
+
+    const auto set_result = resource_limit_object->SetLimitValue(type, static_cast<s64>(value));
+    if (set_result.IsError()) {
+        LOG_ERROR(
+            Kernel_SVC,
+            "Attempted to lower resource limit ({}) for category '{}' below its current value ({})",
+            resource_limit_object->GetMaxResourceValue(type), resource_type,
+            resource_limit_object->GetCurrentResourceValue(type));
+        return set_result;
+    }
+
+    return RESULT_SUCCESS;
+}
+
 namespace {
 struct FunctionDef {
     using Func = void();
@@ -1404,8 +1518,8 @@ static const FunctionDef SVC_Table[] = {
     {0x2D, nullptr, "UnmapPhysicalMemory"},
     {0x2E, nullptr, "GetFutureThreadInfo"},
     {0x2F, nullptr, "GetLastThreadInfo"},
-    {0x30, nullptr, "GetResourceLimitLimitValue"},
-    {0x31, nullptr, "GetResourceLimitCurrentValue"},
+    {0x30, SvcWrap<GetResourceLimitLimitValue>, "GetResourceLimitLimitValue"},
+    {0x31, SvcWrap<GetResourceLimitCurrentValue>, "GetResourceLimitCurrentValue"},
     {0x32, SvcWrap<SetThreadActivity>, "SetThreadActivity"},
     {0x33, SvcWrap<GetThreadContext>, "GetThreadContext"},
     {0x34, SvcWrap<WaitForAddress>, "WaitForAddress"},
@@ -1481,8 +1595,8 @@ static const FunctionDef SVC_Table[] = {
     {0x7A, nullptr, "StartProcess"},
     {0x7B, nullptr, "TerminateProcess"},
     {0x7C, SvcWrap<GetProcessInfo>, "GetProcessInfo"},
-    {0x7D, nullptr, "CreateResourceLimit"},
-    {0x7E, nullptr, "SetResourceLimitLimitValue"},
+    {0x7D, SvcWrap<CreateResourceLimit>, "CreateResourceLimit"},
+    {0x7E, SvcWrap<SetResourceLimitLimitValue>, "SetResourceLimitLimitValue"},
     {0x7F, nullptr, "CallSecureMonitor"},
 };
 

src/core/hle/kernel/svc_wrap.h

@@ -43,6 +43,14 @@ void SvcWrap() {
     FuncReturn(func(static_cast<u32>(Param(0)), static_cast<u32>(Param(1))).raw);
 }
 
+template <ResultCode func(u32*)>
+void SvcWrap() {
+    u32 param = 0;
+    const u32 retval = func(&param).raw;
+    Core::CurrentArmInterface().SetReg(1, param);
+    FuncReturn(retval);
+}
+
 template <ResultCode func(u32*, u32)>
 void SvcWrap() {
     u32 param_1 = 0;

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

@@ -483,11 +483,15 @@ void ICommonStateGetter::GetDefaultDisplayResolution(Kernel::HLERequestContext&
     rb.Push(RESULT_SUCCESS);
 
     if (Settings::values.use_docked_mode) {
-        rb.Push(static_cast<u32>(Service::VI::DisplayResolution::DockedWidth));
-        rb.Push(static_cast<u32>(Service::VI::DisplayResolution::DockedHeight));
+        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));
     } else {
-        rb.Push(static_cast<u32>(Service::VI::DisplayResolution::UndockedWidth));
-        rb.Push(static_cast<u32>(Service::VI::DisplayResolution::UndockedHeight));
+        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));
     }
 
     LOG_DEBUG(Service_AM, "called");

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

@@ -341,6 +341,10 @@ 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");
 
@@ -391,6 +395,7 @@ 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

@@ -49,6 +49,7 @@ 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/interface.cpp

@@ -88,6 +88,20 @@ void NVDRV::FinishInitialize(Kernel::HLERequestContext& ctx) {
     rb.Push(RESULT_SUCCESS);
 }
 
+void NVDRV::GetStatus(Kernel::HLERequestContext& ctx) {
+    LOG_WARNING(Service_NVDRV, "(STUBBED) called");
+    IPC::ResponseBuilder rb{ctx, 2};
+    rb.Push(RESULT_SUCCESS);
+}
+
+void NVDRV::DumpGraphicsMemoryInfo(Kernel::HLERequestContext& ctx) {
+    // According to SwitchBrew, this has no inputs and no outputs, so effectively does nothing on
+    // retail hardware.
+    LOG_DEBUG(Service_NVDRV, "called");
+    IPC::ResponseBuilder rb{ctx, 2};
+    rb.Push(RESULT_SUCCESS);
+}
+
 NVDRV::NVDRV(std::shared_ptr<Module> nvdrv, const char* name)
     : ServiceFramework(name), nvdrv(std::move(nvdrv)) {
     static const FunctionInfo functions[] = {
@@ -97,10 +111,10 @@ NVDRV::NVDRV(std::shared_ptr<Module> nvdrv, const char* name)
         {3, &NVDRV::Initialize, "Initialize"},
         {4, &NVDRV::QueryEvent, "QueryEvent"},
         {5, nullptr, "MapSharedMem"},
-        {6, nullptr, "GetStatus"},
+        {6, &NVDRV::GetStatus, "GetStatus"},
         {7, nullptr, "ForceSetClientPID"},
         {8, &NVDRV::SetClientPID, "SetClientPID"},
-        {9, nullptr, "DumpGraphicsMemoryInfo"},
+        {9, &NVDRV::DumpGraphicsMemoryInfo, "DumpGraphicsMemoryInfo"},
         {10, nullptr, "InitializeDevtools"},
         {11, &NVDRV::Ioctl, "Ioctl2"},
         {12, nullptr, "Ioctl3"},

src/core/hle/service/nvdrv/interface.h

@@ -24,6 +24,8 @@ private:
     void QueryEvent(Kernel::HLERequestContext& ctx);
     void SetClientPID(Kernel::HLERequestContext& ctx);
     void FinishInitialize(Kernel::HLERequestContext& ctx);
+    void GetStatus(Kernel::HLERequestContext& ctx);
+    void DumpGraphicsMemoryInfo(Kernel::HLERequestContext& ctx);
 
     std::shared_ptr<Module> nvdrv;
 

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

@@ -63,6 +63,17 @@ 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) {
 
@@ -127,13 +138,52 @@ 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, nullptr, "RegisterService"},
-        {0x00000003, nullptr, "UnregisterService"},
+        {0x00000002, &SM::RegisterService, "RegisterService"},
+        {0x00000003, &SM::UnregisterService, "UnregisterService"},
     };
     RegisterHandlers(functions);
 }

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

@@ -35,6 +35,8 @@ 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;
 };
@@ -48,6 +50,7 @@ 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

@@ -510,7 +510,11 @@ private:
 
         if (transaction == TransactionId::Connect) {
             IGBPConnectRequestParcel request{ctx.ReadBuffer()};
-            IGBPConnectResponseParcel response{1280, 720};
+            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)};
             ctx.WriteBuffer(response.Serialize());
         } else if (transaction == TransactionId::SetPreallocatedBuffer) {
             IGBPSetPreallocatedBufferRequestParcel request{ctx.ReadBuffer()};
@@ -692,11 +696,15 @@ private:
         rb.Push(RESULT_SUCCESS);
 
         if (Settings::values.use_docked_mode) {
-            rb.Push(static_cast<u32>(Service::VI::DisplayResolution::DockedWidth));
-            rb.Push(static_cast<u32>(Service::VI::DisplayResolution::DockedHeight));
+            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));
         } else {
-            rb.Push(static_cast<u32>(Service::VI::DisplayResolution::UndockedWidth));
-            rb.Push(static_cast<u32>(Service::VI::DisplayResolution::UndockedHeight));
+            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.PushRaw<float>(60.0f);
@@ -901,11 +909,15 @@ private:
         rb.Push(RESULT_SUCCESS);
 
         if (Settings::values.use_docked_mode) {
-            rb.Push(static_cast<u64>(DisplayResolution::DockedWidth));
-            rb.Push(static_cast<u64>(DisplayResolution::DockedHeight));
+            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));
         } else {
-            rb.Push(static_cast<u64>(DisplayResolution::UndockedWidth));
-            rb.Push(static_cast<u64>(DisplayResolution::UndockedHeight));
+            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));
         }
     }
 
@@ -922,6 +934,8 @@ 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

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

src/video_core/CMakeLists.txt

@@ -21,6 +21,8 @@ add_library(video_core STATIC
     macro_interpreter.h
     memory_manager.cpp
     memory_manager.h
+    morton.cpp
+    morton.h
     rasterizer_cache.cpp
     rasterizer_cache.h
     rasterizer_interface.h
@@ -62,7 +64,6 @@ add_library(video_core STATIC
     textures/decoders.cpp
     textures/decoders.h
     textures/texture.h
-    utils.h
     video_core.cpp
     video_core.h
 )

src/video_core/command_processor.cpp

@@ -34,6 +34,9 @@ MICROPROFILE_DEFINE(ProcessCommandLists, "GPU", "Execute command buffer", MP_RGB
 void GPU::ProcessCommandLists(const std::vector<CommandListHeader>& commands) {
     MICROPROFILE_SCOPE(ProcessCommandLists);
 
+    // On entering GPU code, assume all memory may be touched by the ARM core.
+    maxwell_3d->dirty_flags.OnMemoryWrite();
+
     auto WriteReg = [this](u32 method, u32 subchannel, u32 value, u32 remaining_params) {
         LOG_TRACE(HW_GPU,
                   "Processing method {:08X} on subchannel {} value "

src/video_core/engines/fermi_2d.cpp

@@ -2,8 +2,10 @@
 // Licensed under GPLv2 or any later version
 // Refer to the license.txt file included.
 
+#include "core/core.h"
 #include "core/memory.h"
 #include "video_core/engines/fermi_2d.h"
+#include "video_core/engines/maxwell_3d.h"
 #include "video_core/rasterizer_interface.h"
 #include "video_core/textures/decoders.h"
 
@@ -47,6 +49,9 @@ void Fermi2D::HandleSurfaceCopy() {
     u32 dst_bytes_per_pixel = RenderTargetBytesPerPixel(regs.dst.format);
 
     if (!rasterizer.AccelerateSurfaceCopy(regs.src, regs.dst)) {
+        // All copies here update the main memory, so mark all rasterizer states as invalid.
+        Core::System::GetInstance().GPU().Maxwell3D().dirty_flags.OnMemoryWrite();
+
         rasterizer.FlushRegion(source_cpu, src_bytes_per_pixel * regs.src.width * regs.src.height);
         // We have to invalidate the destination region to evict any outdated surfaces from the
         // cache. We do this before actually writing the new data because the destination address

src/video_core/engines/kepler_memory.cpp

@@ -3,8 +3,10 @@
 // Refer to the license.txt file included.
 
 #include "common/logging/log.h"
+#include "core/core.h"
 #include "core/memory.h"
 #include "video_core/engines/kepler_memory.h"
+#include "video_core/engines/maxwell_3d.h"
 #include "video_core/rasterizer_interface.h"
 
 namespace Tegra::Engines {
@@ -47,6 +49,7 @@ void KeplerMemory::ProcessData(u32 data) {
     rasterizer.InvalidateRegion(dest_address, sizeof(u32));
 
     Memory::Write32(dest_address, data);
+    Core::System::GetInstance().GPU().Maxwell3D().dirty_flags.OnMemoryWrite();
 
     state.write_offset++;
 }

src/video_core/engines/maxwell_3d.cpp

@@ -135,10 +135,24 @@ void Maxwell3D::WriteReg(u32 method, u32 value, u32 remaining_params) {
 
     if (regs.reg_array[method] != value) {
         regs.reg_array[method] = value;
+        // Vertex format
         if (method >= MAXWELL3D_REG_INDEX(vertex_attrib_format) &&
             method < MAXWELL3D_REG_INDEX(vertex_attrib_format) + regs.vertex_attrib_format.size()) {
             dirty_flags.vertex_attrib_format = true;
         }
+
+        // Vertex buffer
+        if (method >= MAXWELL3D_REG_INDEX(vertex_array) &&
+            method < MAXWELL3D_REG_INDEX(vertex_array) + 4 * 32) {
+            dirty_flags.vertex_array |= 1u << ((method - MAXWELL3D_REG_INDEX(vertex_array)) >> 2);
+        } else if (method >= MAXWELL3D_REG_INDEX(vertex_array_limit) &&
+                   method < MAXWELL3D_REG_INDEX(vertex_array_limit) + 2 * 32) {
+            dirty_flags.vertex_array |=
+                1u << ((method - MAXWELL3D_REG_INDEX(vertex_array_limit)) >> 1);
+        } else if (method >= MAXWELL3D_REG_INDEX(instanced_arrays) &&
+                   method < MAXWELL3D_REG_INDEX(instanced_arrays) + 32) {
+            dirty_flags.vertex_array |= 1u << (method - MAXWELL3D_REG_INDEX(instanced_arrays));
+        }
     }
 
     switch (method) {
@@ -270,6 +284,7 @@ void Maxwell3D::ProcessQueryGet() {
             query_result.timestamp = CoreTiming::GetTicks();
             Memory::WriteBlock(*address, &query_result, sizeof(query_result));
         }
+        dirty_flags.OnMemoryWrite();
         break;
     }
     default:
@@ -346,6 +361,7 @@ void Maxwell3D::ProcessCBData(u32 value) {
         memory_manager.GpuToCpuAddress(buffer_address + regs.const_buffer.cb_pos);
 
     Memory::Write32(*address, value);
+    dirty_flags.OnMemoryWrite();
 
     // Increment the current buffer position.
     regs.const_buffer.cb_pos = regs.const_buffer.cb_pos + 4;

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;
 
@@ -631,7 +639,16 @@ public:
                     }
                 } zeta;
 
-                INSERT_PADDING_WORDS(0x5B);
+                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);
 
                 std::array<VertexAttribute, NumVertexAttributes> vertex_attrib_format;
 
@@ -719,6 +736,7 @@ public:
                 u32 frag_color_clamp;
 
                 union {
+                    BitField<0, 1, u32> y_negate;
                     BitField<4, 1, u32> triangle_rast_flip;
                 } screen_y_control;
 
@@ -752,7 +770,11 @@ public:
                     }
                 } tsc;
 
-                INSERT_PADDING_WORDS(0x3);
+                INSERT_PADDING_WORDS(0x1);
+
+                float polygon_offset_factor;
+
+                INSERT_PADDING_WORDS(0x1);
 
                 struct {
                     u32 tic_address_high;
@@ -777,7 +799,9 @@ public:
 
                 u32 framebuffer_srgb;
 
-                INSERT_PADDING_WORDS(0x12);
+                float polygon_offset_units;
+
+                INSERT_PADDING_WORDS(0x11);
 
                 union {
                     BitField<2, 1, u32> coord_origin;
@@ -854,7 +878,9 @@ public:
 
                 INSERT_PADDING_WORDS(0x7);
 
-                INSERT_PADDING_WORDS(0x20);
+                INSERT_PADDING_WORDS(0x1F);
+
+                float polygon_offset_clamp;
 
                 struct {
                     u32 is_instanced[NumVertexArrays];
@@ -870,7 +896,13 @@ public:
 
                 Cull cull;
 
-                INSERT_PADDING_WORDS(0x28);
+                u32 pixel_center_integer;
+
+                INSERT_PADDING_WORDS(0x1);
+
+                u32 viewport_transform_enabled;
+
+                INSERT_PADDING_WORDS(0x25);
 
                 struct {
                     u32 enable;
@@ -1035,6 +1067,11 @@ public:
 
     struct DirtyFlags {
         bool vertex_attrib_format = true;
+        u32 vertex_array = 0xFFFFFFFF;
+
+        void OnMemoryWrite() {
+            vertex_array = 0xFFFFFFFF;
+        }
     };
 
     DirtyFlags dirty_flags;
@@ -1127,6 +1164,9 @@ 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);
@@ -1134,6 +1174,7 @@ 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);
@@ -1164,6 +1205,7 @@ 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);
@@ -1171,13 +1213,17 @@ 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(pixel_center_integer, 0x649);
+ASSERT_REG_POSITION(viewport_transform_enabled, 0x64B);
 ASSERT_REG_POSITION(logic_op, 0x671);
 ASSERT_REG_POSITION(clear_buffers, 0x674);
 ASSERT_REG_POSITION(color_mask, 0x680);

src/video_core/engines/maxwell_dma.cpp

@@ -2,7 +2,9 @@
 // Licensed under GPLv2 or any later version
 // Refer to the license.txt file included.
 
+#include "core/core.h"
 #include "core/memory.h"
+#include "video_core/engines/maxwell_3d.h"
 #include "video_core/engines/maxwell_dma.h"
 #include "video_core/rasterizer_interface.h"
 #include "video_core/textures/decoders.h"
@@ -54,6 +56,9 @@ void MaxwellDMA::HandleCopy() {
         return;
     }
 
+    // All copies here update the main memory, so mark all rasterizer states as invalid.
+    Core::System::GetInstance().GPU().Maxwell3D().dirty_flags.OnMemoryWrite();
+
     if (regs.exec.is_dst_linear && regs.exec.is_src_linear) {
         // When the enable_2d bit is disabled, the copy is performed as if we were copying a 1D
         // buffer of length `x_count`, otherwise we copy a 2D image of dimensions (x_count,

src/video_core/engines/shader_bytecode.h

@@ -82,6 +82,8 @@ union Attribute {
         Position = 7,
         Attribute_0 = 8,
         Attribute_31 = 39,
+        ClipDistances0123 = 44,
+        ClipDistances4567 = 45,
         PointCoord = 46,
         // This attribute contains a tuple of (~, ~, InstanceId, VertexId) when inside a vertex
         // shader, and a tuple of (TessCoord.x, TessCoord.y, TessCoord.z, ~) when inside a Tess Eval
@@ -262,7 +264,7 @@ enum class FlowCondition : u64 {
     Fcsm_Tr = 0x1C, // TODO(bunnei): What is this used for?
 };
 
-enum class ControlCode : u64 {
+enum class ConditionCode : u64 {
     F = 0,
     LT = 1,
     EQ = 2,
@@ -366,6 +368,11 @@ enum class HalfPrecision : u64 {
     FMZ = 2,
 };
 
+enum class R2pMode : u64 {
+    Pr = 0,
+    Cc = 1,
+};
+
 enum class IpaInterpMode : u64 {
     Linear = 0,
     Perspective = 1,
@@ -570,7 +577,6 @@ 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;
 
@@ -832,7 +838,7 @@ union Instruction {
     union {
         BitField<0, 3, u64> pred0;
         BitField<3, 3, u64> pred3;
-        BitField<8, 5, ControlCode> cc; // flag in cc
+        BitField<8, 5, ConditionCode> cc; // flag in cc
         BitField<39, 3, u64> pred39;
         BitField<42, 1, u64> neg_pred39;
         BitField<45, 4, PredOperation> op; // op with pred39
@@ -855,6 +861,12 @@ union Instruction {
         BitField<39, 3, u64> pred39;
     } hsetp2;
 
+    union {
+        BitField<40, 1, R2pMode> mode;
+        BitField<41, 2, u64> byte;
+        BitField<20, 7, u64> immediate_mask;
+    } r2p;
+
     union {
         BitField<39, 3, u64> pred39;
         BitField<42, 1, u64> neg_pred;
@@ -1236,7 +1248,7 @@ 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, ControlCode> flow_control_code;
+    BitField<0, 5, ConditionCode> flow_condition_code;
 
     Attribute attribute;
     Sampler sampler;
@@ -1257,6 +1269,7 @@ public:
         BFE_C,
         BFE_R,
         BFE_IMM,
+        BFI_IMM_R,
         BRA,
         PBK,
         LD_A,
@@ -1382,6 +1395,7 @@ public:
         PSETP,
         PSET,
         CSETP,
+        R2P_IMM,
         XMAD_IMM,
         XMAD_CR,
         XMAD_RC,
@@ -1397,6 +1411,7 @@ public:
         ArithmeticHalf,
         ArithmeticHalfImmediate,
         Bfe,
+        Bfi,
         Shift,
         Ffma,
         Hfma2,
@@ -1411,6 +1426,7 @@ public:
         HalfSetPredicate,
         PredicateSetPredicate,
         PredicateSetRegister,
+        RegisterSetPredicate,
         Conversion,
         Xmad,
         Unknown,
@@ -1614,6 +1630,7 @@ private:
             INST("0100110000000---", Id::BFE_C, Type::Bfe, "BFE_C"),
             INST("0101110000000---", Id::BFE_R, Type::Bfe, "BFE_R"),
             INST("0011100-00000---", Id::BFE_IMM, Type::Bfe, "BFE_IMM"),
+            INST("0011011-11110---", Id::BFI_IMM_R, Type::Bfi, "BFI_IMM_R"),
             INST("0100110001000---", Id::LOP_C, Type::ArithmeticInteger, "LOP_C"),
             INST("0101110001000---", Id::LOP_R, Type::ArithmeticInteger, "LOP_R"),
             INST("0011100001000---", Id::LOP_IMM, Type::ArithmeticInteger, "LOP_IMM"),
@@ -1648,6 +1665,7 @@ private:
             INST("0101000010001---", Id::PSET, Type::PredicateSetRegister, "PSET"),
             INST("0101000010010---", Id::PSETP, Type::PredicateSetPredicate, "PSETP"),
             INST("010100001010----", Id::CSETP, Type::PredicateSetPredicate, "CSETP"),
+            INST("0011100-11110---", Id::R2P_IMM, Type::RegisterSetPredicate, "R2P_IMM"),
             INST("0011011-00------", Id::XMAD_IMM, Type::Xmad, "XMAD_IMM"),
             INST("0100111---------", Id::XMAD_CR, Type::Xmad, "XMAD_CR"),
             INST("010100010-------", Id::XMAD_RC, Type::Xmad, "XMAD_RC"),

src/video_core/engines/shader_header.h

@@ -62,7 +62,16 @@ struct Header {
             INSERT_PADDING_BYTES(1);  // ImapSystemValuesB
             INSERT_PADDING_BYTES(16); // ImapGenericVector[32]
             INSERT_PADDING_BYTES(2);  // ImapColor
-            INSERT_PADDING_BYTES(2);  // ImapSystemValuesC
+            union {
+                BitField<0, 8, u16> clip_distances;
+                BitField<8, 1, u16> point_sprite_s;
+                BitField<9, 1, u16> point_sprite_t;
+                BitField<10, 1, u16> fog_coordinate;
+                BitField<12, 1, u16> tessellation_eval_point_u;
+                BitField<13, 1, u16> tessellation_eval_point_v;
+                BitField<14, 1, u16> instance_id;
+                BitField<15, 1, u16> vertex_id;
+            };
             INSERT_PADDING_BYTES(5);  // ImapFixedFncTexture[10]
             INSERT_PADDING_BYTES(1);  // ImapReserved
             INSERT_PADDING_BYTES(3);  // OmapSystemValuesA

src/video_core/morton.cpp

@@ -0,0 +1,354 @@
+// Copyright 2018 yuzu Emulator Project
+// Licensed under GPLv2 or any later version
+// Refer to the license.txt file included.
+
+#include <array>
+#include <cstring>
+#include "common/assert.h"
+#include "common/common_types.h"
+#include "core/memory.h"
+#include "video_core/morton.h"
+#include "video_core/surface.h"
+#include "video_core/textures/decoders.h"
+
+namespace VideoCore {
+
+using Surface::GetBytesPerPixel;
+using Surface::PixelFormat;
+
+using MortonCopyFn = void (*)(u32, u32, u32, u32, u32, u8*, std::size_t, VAddr);
+using ConversionArray = std::array<MortonCopyFn, Surface::MaxPixelFormat>;
+
+template <bool morton_to_linear, PixelFormat format>
+static void MortonCopy(u32 stride, u32 block_height, u32 height, u32 block_depth, u32 depth,
+                       u8* buffer, std::size_t buffer_size, VAddr addr) {
+    constexpr u32 bytes_per_pixel = GetBytesPerPixel(format);
+
+    // With the BCn formats (DXT and DXN), each 4x4 tile is swizzled instead of just individual
+    // pixel values.
+    const u32 tile_size_x{GetDefaultBlockWidth(format)};
+    const u32 tile_size_y{GetDefaultBlockHeight(format)};
+
+    if constexpr (morton_to_linear) {
+        Tegra::Texture::UnswizzleTexture(buffer, addr, tile_size_x, tile_size_y, bytes_per_pixel,
+                                         stride, height, depth, block_height, block_depth);
+    } else {
+        Tegra::Texture::CopySwizzledData((stride + tile_size_x - 1) / tile_size_x,
+                                         (height + tile_size_y - 1) / tile_size_y, depth,
+                                         bytes_per_pixel, bytes_per_pixel, Memory::GetPointer(addr),
+                                         buffer, false, block_height, block_depth);
+    }
+}
+
+static constexpr ConversionArray morton_to_linear_fns = {
+    // clang-format off
+        MortonCopy<true, PixelFormat::ABGR8U>,
+        MortonCopy<true, PixelFormat::ABGR8S>,
+        MortonCopy<true, PixelFormat::ABGR8UI>,
+        MortonCopy<true, PixelFormat::B5G6R5U>,
+        MortonCopy<true, PixelFormat::A2B10G10R10U>,
+        MortonCopy<true, PixelFormat::A1B5G5R5U>,
+        MortonCopy<true, PixelFormat::R8U>,
+        MortonCopy<true, PixelFormat::R8UI>,
+        MortonCopy<true, PixelFormat::RGBA16F>,
+        MortonCopy<true, PixelFormat::RGBA16U>,
+        MortonCopy<true, PixelFormat::RGBA16UI>,
+        MortonCopy<true, PixelFormat::R11FG11FB10F>,
+        MortonCopy<true, PixelFormat::RGBA32UI>,
+        MortonCopy<true, PixelFormat::DXT1>,
+        MortonCopy<true, PixelFormat::DXT23>,
+        MortonCopy<true, PixelFormat::DXT45>,
+        MortonCopy<true, PixelFormat::DXN1>,
+        MortonCopy<true, PixelFormat::DXN2UNORM>,
+        MortonCopy<true, PixelFormat::DXN2SNORM>,
+        MortonCopy<true, PixelFormat::BC7U>,
+        MortonCopy<true, PixelFormat::BC6H_UF16>,
+        MortonCopy<true, PixelFormat::BC6H_SF16>,
+        MortonCopy<true, PixelFormat::ASTC_2D_4X4>,
+        MortonCopy<true, PixelFormat::G8R8U>,
+        MortonCopy<true, PixelFormat::G8R8S>,
+        MortonCopy<true, PixelFormat::BGRA8>,
+        MortonCopy<true, PixelFormat::RGBA32F>,
+        MortonCopy<true, PixelFormat::RG32F>,
+        MortonCopy<true, PixelFormat::R32F>,
+        MortonCopy<true, PixelFormat::R16F>,
+        MortonCopy<true, PixelFormat::R16U>,
+        MortonCopy<true, PixelFormat::R16S>,
+        MortonCopy<true, PixelFormat::R16UI>,
+        MortonCopy<true, PixelFormat::R16I>,
+        MortonCopy<true, PixelFormat::RG16>,
+        MortonCopy<true, PixelFormat::RG16F>,
+        MortonCopy<true, PixelFormat::RG16UI>,
+        MortonCopy<true, PixelFormat::RG16I>,
+        MortonCopy<true, PixelFormat::RG16S>,
+        MortonCopy<true, PixelFormat::RGB32F>,
+        MortonCopy<true, PixelFormat::RGBA8_SRGB>,
+        MortonCopy<true, PixelFormat::RG8U>,
+        MortonCopy<true, PixelFormat::RG8S>,
+        MortonCopy<true, PixelFormat::RG32UI>,
+        MortonCopy<true, PixelFormat::R32UI>,
+        MortonCopy<true, PixelFormat::ASTC_2D_8X8>,
+        MortonCopy<true, PixelFormat::ASTC_2D_8X5>,
+        MortonCopy<true, PixelFormat::ASTC_2D_5X4>,
+        MortonCopy<true, PixelFormat::BGRA8_SRGB>,
+        MortonCopy<true, PixelFormat::DXT1_SRGB>,
+        MortonCopy<true, PixelFormat::DXT23_SRGB>,
+        MortonCopy<true, PixelFormat::DXT45_SRGB>,
+        MortonCopy<true, PixelFormat::BC7U_SRGB>,
+        MortonCopy<true, PixelFormat::ASTC_2D_4X4_SRGB>,
+        MortonCopy<true, PixelFormat::ASTC_2D_8X8_SRGB>,
+        MortonCopy<true, PixelFormat::ASTC_2D_8X5_SRGB>,
+        MortonCopy<true, PixelFormat::ASTC_2D_5X4_SRGB>,
+        MortonCopy<true, PixelFormat::ASTC_2D_5X5>,
+        MortonCopy<true, PixelFormat::ASTC_2D_5X5_SRGB>,
+        MortonCopy<true, PixelFormat::ASTC_2D_10X8>,
+        MortonCopy<true, PixelFormat::ASTC_2D_10X8_SRGB>,
+        MortonCopy<true, PixelFormat::Z32F>,
+        MortonCopy<true, PixelFormat::Z16>,
+        MortonCopy<true, PixelFormat::Z24S8>,
+        MortonCopy<true, PixelFormat::S8Z24>,
+        MortonCopy<true, PixelFormat::Z32FS8>,
+    // clang-format on
+};
+
+static constexpr ConversionArray linear_to_morton_fns = {
+    // clang-format off
+        MortonCopy<false, PixelFormat::ABGR8U>,
+        MortonCopy<false, PixelFormat::ABGR8S>,
+        MortonCopy<false, PixelFormat::ABGR8UI>,
+        MortonCopy<false, PixelFormat::B5G6R5U>,
+        MortonCopy<false, PixelFormat::A2B10G10R10U>,
+        MortonCopy<false, PixelFormat::A1B5G5R5U>,
+        MortonCopy<false, PixelFormat::R8U>,
+        MortonCopy<false, PixelFormat::R8UI>,
+        MortonCopy<false, PixelFormat::RGBA16F>,
+        MortonCopy<false, PixelFormat::RGBA16U>,
+        MortonCopy<false, PixelFormat::RGBA16UI>,
+        MortonCopy<false, PixelFormat::R11FG11FB10F>,
+        MortonCopy<false, PixelFormat::RGBA32UI>,
+        MortonCopy<false, PixelFormat::DXT1>,
+        MortonCopy<false, PixelFormat::DXT23>,
+        MortonCopy<false, PixelFormat::DXT45>,
+        MortonCopy<false, PixelFormat::DXN1>,
+        MortonCopy<false, PixelFormat::DXN2UNORM>,
+        MortonCopy<false, PixelFormat::DXN2SNORM>,
+        MortonCopy<false, PixelFormat::BC7U>,
+        MortonCopy<false, PixelFormat::BC6H_UF16>,
+        MortonCopy<false, PixelFormat::BC6H_SF16>,
+        // TODO(Subv): Swizzling ASTC formats are not supported
+        nullptr,
+        MortonCopy<false, PixelFormat::G8R8U>,
+        MortonCopy<false, PixelFormat::G8R8S>,
+        MortonCopy<false, PixelFormat::BGRA8>,
+        MortonCopy<false, PixelFormat::RGBA32F>,
+        MortonCopy<false, PixelFormat::RG32F>,
+        MortonCopy<false, PixelFormat::R32F>,
+        MortonCopy<false, PixelFormat::R16F>,
+        MortonCopy<false, PixelFormat::R16U>,
+        MortonCopy<false, PixelFormat::R16S>,
+        MortonCopy<false, PixelFormat::R16UI>,
+        MortonCopy<false, PixelFormat::R16I>,
+        MortonCopy<false, PixelFormat::RG16>,
+        MortonCopy<false, PixelFormat::RG16F>,
+        MortonCopy<false, PixelFormat::RG16UI>,
+        MortonCopy<false, PixelFormat::RG16I>,
+        MortonCopy<false, PixelFormat::RG16S>,
+        MortonCopy<false, PixelFormat::RGB32F>,
+        MortonCopy<false, PixelFormat::RGBA8_SRGB>,
+        MortonCopy<false, PixelFormat::RG8U>,
+        MortonCopy<false, PixelFormat::RG8S>,
+        MortonCopy<false, PixelFormat::RG32UI>,
+        MortonCopy<false, PixelFormat::R32UI>,
+        nullptr,
+        nullptr,
+        nullptr,
+        MortonCopy<false, PixelFormat::BGRA8_SRGB>,
+        MortonCopy<false, PixelFormat::DXT1_SRGB>,
+        MortonCopy<false, PixelFormat::DXT23_SRGB>,
+        MortonCopy<false, PixelFormat::DXT45_SRGB>,
+        MortonCopy<false, PixelFormat::BC7U_SRGB>,
+        nullptr,
+        nullptr,
+        nullptr,
+        nullptr,
+        nullptr,
+        nullptr,
+        nullptr,
+        nullptr,
+        MortonCopy<false, PixelFormat::Z32F>,
+        MortonCopy<false, PixelFormat::Z16>,
+        MortonCopy<false, PixelFormat::Z24S8>,
+        MortonCopy<false, PixelFormat::S8Z24>,
+        MortonCopy<false, PixelFormat::Z32FS8>,
+    // clang-format on
+};
+
+static MortonCopyFn GetSwizzleFunction(MortonSwizzleMode mode, Surface::PixelFormat format) {
+    switch (mode) {
+    case MortonSwizzleMode::MortonToLinear:
+        return morton_to_linear_fns[static_cast<std::size_t>(format)];
+    case MortonSwizzleMode::LinearToMorton:
+        return linear_to_morton_fns[static_cast<std::size_t>(format)];
+    }
+    UNREACHABLE();
+}
+
+/// 8x8 Z-Order coordinate from 2D coordinates
+static u32 MortonInterleave(u32 x, u32 y) {
+    static const u32 xlut[] = {0x00, 0x01, 0x04, 0x05, 0x10, 0x11, 0x14, 0x15};
+    static const u32 ylut[] = {0x00, 0x02, 0x08, 0x0a, 0x20, 0x22, 0x28, 0x2a};
+    return xlut[x % 8] + ylut[y % 8];
+}
+
+/// Calculates the offset of the position of the pixel in Morton order
+static u32 GetMortonOffset(u32 x, u32 y, u32 bytes_per_pixel) {
+    // Images are split into 8x8 tiles. Each tile is composed of four 4x4 subtiles each
+    // of which is composed of four 2x2 subtiles each of which is composed of four texels.
+    // Each structure is embedded into the next-bigger one in a diagonal pattern, e.g.
+    // texels are laid out in a 2x2 subtile like this:
+    // 2 3
+    // 0 1
+    //
+    // The full 8x8 tile has the texels arranged like this:
+    //
+    // 42 43 46 47 58 59 62 63
+    // 40 41 44 45 56 57 60 61
+    // 34 35 38 39 50 51 54 55
+    // 32 33 36 37 48 49 52 53
+    // 10 11 14 15 26 27 30 31
+    // 08 09 12 13 24 25 28 29
+    // 02 03 06 07 18 19 22 23
+    // 00 01 04 05 16 17 20 21
+    //
+    // This pattern is what's called Z-order curve, or Morton order.
+
+    const unsigned int block_height = 8;
+    const unsigned int coarse_x = x & ~7;
+
+    u32 i = MortonInterleave(x, y);
+
+    const unsigned int offset = coarse_x * block_height;
+
+    return (i + offset) * bytes_per_pixel;
+}
+
+static u32 MortonInterleave128(u32 x, u32 y) {
+    // 128x128 Z-Order coordinate from 2D coordinates
+    static constexpr u32 xlut[] = {
+        0x0000, 0x0001, 0x0002, 0x0003, 0x0008, 0x0009, 0x000a, 0x000b, 0x0040, 0x0041, 0x0042,
+        0x0043, 0x0048, 0x0049, 0x004a, 0x004b, 0x0800, 0x0801, 0x0802, 0x0803, 0x0808, 0x0809,
+        0x080a, 0x080b, 0x0840, 0x0841, 0x0842, 0x0843, 0x0848, 0x0849, 0x084a, 0x084b, 0x1000,
+        0x1001, 0x1002, 0x1003, 0x1008, 0x1009, 0x100a, 0x100b, 0x1040, 0x1041, 0x1042, 0x1043,
+        0x1048, 0x1049, 0x104a, 0x104b, 0x1800, 0x1801, 0x1802, 0x1803, 0x1808, 0x1809, 0x180a,
+        0x180b, 0x1840, 0x1841, 0x1842, 0x1843, 0x1848, 0x1849, 0x184a, 0x184b, 0x2000, 0x2001,
+        0x2002, 0x2003, 0x2008, 0x2009, 0x200a, 0x200b, 0x2040, 0x2041, 0x2042, 0x2043, 0x2048,
+        0x2049, 0x204a, 0x204b, 0x2800, 0x2801, 0x2802, 0x2803, 0x2808, 0x2809, 0x280a, 0x280b,
+        0x2840, 0x2841, 0x2842, 0x2843, 0x2848, 0x2849, 0x284a, 0x284b, 0x3000, 0x3001, 0x3002,
+        0x3003, 0x3008, 0x3009, 0x300a, 0x300b, 0x3040, 0x3041, 0x3042, 0x3043, 0x3048, 0x3049,
+        0x304a, 0x304b, 0x3800, 0x3801, 0x3802, 0x3803, 0x3808, 0x3809, 0x380a, 0x380b, 0x3840,
+        0x3841, 0x3842, 0x3843, 0x3848, 0x3849, 0x384a, 0x384b, 0x0000, 0x0001, 0x0002, 0x0003,
+        0x0008, 0x0009, 0x000a, 0x000b, 0x0040, 0x0041, 0x0042, 0x0043, 0x0048, 0x0049, 0x004a,
+        0x004b, 0x0800, 0x0801, 0x0802, 0x0803, 0x0808, 0x0809, 0x080a, 0x080b, 0x0840, 0x0841,
+        0x0842, 0x0843, 0x0848, 0x0849, 0x084a, 0x084b, 0x1000, 0x1001, 0x1002, 0x1003, 0x1008,
+        0x1009, 0x100a, 0x100b, 0x1040, 0x1041, 0x1042, 0x1043, 0x1048, 0x1049, 0x104a, 0x104b,
+        0x1800, 0x1801, 0x1802, 0x1803, 0x1808, 0x1809, 0x180a, 0x180b, 0x1840, 0x1841, 0x1842,
+        0x1843, 0x1848, 0x1849, 0x184a, 0x184b, 0x2000, 0x2001, 0x2002, 0x2003, 0x2008, 0x2009,
+        0x200a, 0x200b, 0x2040, 0x2041, 0x2042, 0x2043, 0x2048, 0x2049, 0x204a, 0x204b, 0x2800,
+        0x2801, 0x2802, 0x2803, 0x2808, 0x2809, 0x280a, 0x280b, 0x2840, 0x2841, 0x2842, 0x2843,
+        0x2848, 0x2849, 0x284a, 0x284b, 0x3000, 0x3001, 0x3002, 0x3003, 0x3008, 0x3009, 0x300a,
+        0x300b, 0x3040, 0x3041, 0x3042, 0x3043, 0x3048, 0x3049, 0x304a, 0x304b, 0x3800, 0x3801,
+        0x3802, 0x3803, 0x3808, 0x3809, 0x380a, 0x380b, 0x3840, 0x3841, 0x3842, 0x3843, 0x3848,
+        0x3849, 0x384a, 0x384b, 0x0000, 0x0001, 0x0002, 0x0003, 0x0008, 0x0009, 0x000a, 0x000b,
+        0x0040, 0x0041, 0x0042, 0x0043, 0x0048, 0x0049, 0x004a, 0x004b, 0x0800, 0x0801, 0x0802,
+        0x0803, 0x0808, 0x0809, 0x080a, 0x080b, 0x0840, 0x0841, 0x0842, 0x0843, 0x0848, 0x0849,
+        0x084a, 0x084b, 0x1000, 0x1001, 0x1002, 0x1003, 0x1008, 0x1009, 0x100a, 0x100b, 0x1040,
+        0x1041, 0x1042, 0x1043, 0x1048, 0x1049, 0x104a, 0x104b, 0x1800, 0x1801, 0x1802, 0x1803,
+        0x1808, 0x1809, 0x180a, 0x180b, 0x1840, 0x1841, 0x1842, 0x1843, 0x1848, 0x1849, 0x184a,
+        0x184b, 0x2000, 0x2001, 0x2002, 0x2003, 0x2008, 0x2009, 0x200a, 0x200b, 0x2040, 0x2041,
+        0x2042, 0x2043, 0x2048, 0x2049, 0x204a, 0x204b, 0x2800, 0x2801, 0x2802, 0x2803, 0x2808,
+        0x2809, 0x280a, 0x280b, 0x2840, 0x2841, 0x2842, 0x2843, 0x2848, 0x2849, 0x284a, 0x284b,
+        0x3000, 0x3001, 0x3002, 0x3003, 0x3008, 0x3009, 0x300a, 0x300b, 0x3040, 0x3041, 0x3042,
+        0x3043, 0x3048, 0x3049, 0x304a, 0x304b, 0x3800, 0x3801, 0x3802, 0x3803, 0x3808, 0x3809,
+        0x380a, 0x380b, 0x3840, 0x3841, 0x3842, 0x3843, 0x3848, 0x3849, 0x384a, 0x384b,
+    };
+    static constexpr u32 ylut[] = {
+        0x0000, 0x0004, 0x0010, 0x0014, 0x0020, 0x0024, 0x0030, 0x0034, 0x0080, 0x0084, 0x0090,
+        0x0094, 0x00a0, 0x00a4, 0x00b0, 0x00b4, 0x0100, 0x0104, 0x0110, 0x0114, 0x0120, 0x0124,
+        0x0130, 0x0134, 0x0180, 0x0184, 0x0190, 0x0194, 0x01a0, 0x01a4, 0x01b0, 0x01b4, 0x0200,
+        0x0204, 0x0210, 0x0214, 0x0220, 0x0224, 0x0230, 0x0234, 0x0280, 0x0284, 0x0290, 0x0294,
+        0x02a0, 0x02a4, 0x02b0, 0x02b4, 0x0300, 0x0304, 0x0310, 0x0314, 0x0320, 0x0324, 0x0330,
+        0x0334, 0x0380, 0x0384, 0x0390, 0x0394, 0x03a0, 0x03a4, 0x03b0, 0x03b4, 0x0400, 0x0404,
+        0x0410, 0x0414, 0x0420, 0x0424, 0x0430, 0x0434, 0x0480, 0x0484, 0x0490, 0x0494, 0x04a0,
+        0x04a4, 0x04b0, 0x04b4, 0x0500, 0x0504, 0x0510, 0x0514, 0x0520, 0x0524, 0x0530, 0x0534,
+        0x0580, 0x0584, 0x0590, 0x0594, 0x05a0, 0x05a4, 0x05b0, 0x05b4, 0x0600, 0x0604, 0x0610,
+        0x0614, 0x0620, 0x0624, 0x0630, 0x0634, 0x0680, 0x0684, 0x0690, 0x0694, 0x06a0, 0x06a4,
+        0x06b0, 0x06b4, 0x0700, 0x0704, 0x0710, 0x0714, 0x0720, 0x0724, 0x0730, 0x0734, 0x0780,
+        0x0784, 0x0790, 0x0794, 0x07a0, 0x07a4, 0x07b0, 0x07b4, 0x0000, 0x0004, 0x0010, 0x0014,
+        0x0020, 0x0024, 0x0030, 0x0034, 0x0080, 0x0084, 0x0090, 0x0094, 0x00a0, 0x00a4, 0x00b0,
+        0x00b4, 0x0100, 0x0104, 0x0110, 0x0114, 0x0120, 0x0124, 0x0130, 0x0134, 0x0180, 0x0184,
+        0x0190, 0x0194, 0x01a0, 0x01a4, 0x01b0, 0x01b4, 0x0200, 0x0204, 0x0210, 0x0214, 0x0220,
+        0x0224, 0x0230, 0x0234, 0x0280, 0x0284, 0x0290, 0x0294, 0x02a0, 0x02a4, 0x02b0, 0x02b4,
+        0x0300, 0x0304, 0x0310, 0x0314, 0x0320, 0x0324, 0x0330, 0x0334, 0x0380, 0x0384, 0x0390,
+        0x0394, 0x03a0, 0x03a4, 0x03b0, 0x03b4, 0x0400, 0x0404, 0x0410, 0x0414, 0x0420, 0x0424,
+        0x0430, 0x0434, 0x0480, 0x0484, 0x0490, 0x0494, 0x04a0, 0x04a4, 0x04b0, 0x04b4, 0x0500,
+        0x0504, 0x0510, 0x0514, 0x0520, 0x0524, 0x0530, 0x0534, 0x0580, 0x0584, 0x0590, 0x0594,
+        0x05a0, 0x05a4, 0x05b0, 0x05b4, 0x0600, 0x0604, 0x0610, 0x0614, 0x0620, 0x0624, 0x0630,
+        0x0634, 0x0680, 0x0684, 0x0690, 0x0694, 0x06a0, 0x06a4, 0x06b0, 0x06b4, 0x0700, 0x0704,
+        0x0710, 0x0714, 0x0720, 0x0724, 0x0730, 0x0734, 0x0780, 0x0784, 0x0790, 0x0794, 0x07a0,
+        0x07a4, 0x07b0, 0x07b4, 0x0000, 0x0004, 0x0010, 0x0014, 0x0020, 0x0024, 0x0030, 0x0034,
+        0x0080, 0x0084, 0x0090, 0x0094, 0x00a0, 0x00a4, 0x00b0, 0x00b4, 0x0100, 0x0104, 0x0110,
+        0x0114, 0x0120, 0x0124, 0x0130, 0x0134, 0x0180, 0x0184, 0x0190, 0x0194, 0x01a0, 0x01a4,
+        0x01b0, 0x01b4, 0x0200, 0x0204, 0x0210, 0x0214, 0x0220, 0x0224, 0x0230, 0x0234, 0x0280,
+        0x0284, 0x0290, 0x0294, 0x02a0, 0x02a4, 0x02b0, 0x02b4, 0x0300, 0x0304, 0x0310, 0x0314,
+        0x0320, 0x0324, 0x0330, 0x0334, 0x0380, 0x0384, 0x0390, 0x0394, 0x03a0, 0x03a4, 0x03b0,
+        0x03b4, 0x0400, 0x0404, 0x0410, 0x0414, 0x0420, 0x0424, 0x0430, 0x0434, 0x0480, 0x0484,
+        0x0490, 0x0494, 0x04a0, 0x04a4, 0x04b0, 0x04b4, 0x0500, 0x0504, 0x0510, 0x0514, 0x0520,
+        0x0524, 0x0530, 0x0534, 0x0580, 0x0584, 0x0590, 0x0594, 0x05a0, 0x05a4, 0x05b0, 0x05b4,
+        0x0600, 0x0604, 0x0610, 0x0614, 0x0620, 0x0624, 0x0630, 0x0634, 0x0680, 0x0684, 0x0690,
+        0x0694, 0x06a0, 0x06a4, 0x06b0, 0x06b4, 0x0700, 0x0704, 0x0710, 0x0714, 0x0720, 0x0724,
+        0x0730, 0x0734, 0x0780, 0x0784, 0x0790, 0x0794, 0x07a0, 0x07a4, 0x07b0, 0x07b4,
+    };
+    return xlut[x % 128] + ylut[y % 128];
+}
+
+static u32 GetMortonOffset128(u32 x, u32 y, u32 bytes_per_pixel) {
+    // Calculates the offset of the position of the pixel in Morton order
+    // Framebuffer images are split into 128x128 tiles.
+
+    constexpr u32 block_height = 128;
+    const u32 coarse_x = x & ~127;
+
+    const u32 i = MortonInterleave128(x, y);
+
+    const u32 offset = coarse_x * block_height;
+
+    return (i + offset) * bytes_per_pixel;
+}
+
+void MortonSwizzle(MortonSwizzleMode mode, Surface::PixelFormat format, u32 stride,
+                   u32 block_height, u32 height, u32 block_depth, u32 depth, u8* buffer,
+                   std::size_t buffer_size, VAddr addr) {
+
+    GetSwizzleFunction(mode, format)(stride, block_height, height, block_depth, depth, buffer,
+                                     buffer_size, addr);
+}
+
+void MortonCopyPixels128(u32 width, u32 height, u32 bytes_per_pixel, u32 linear_bytes_per_pixel,
+                         u8* morton_data, u8* linear_data, bool morton_to_linear) {
+    u8* data_ptrs[2];
+    for (u32 y = 0; y < height; ++y) {
+        for (u32 x = 0; x < width; ++x) {
+            const u32 coarse_y = y & ~127;
+            const u32 morton_offset =
+                GetMortonOffset128(x, y, bytes_per_pixel) + coarse_y * width * bytes_per_pixel;
+            const u32 linear_pixel_index = (x + y * width) * linear_bytes_per_pixel;
+
+            data_ptrs[morton_to_linear ? 1 : 0] = morton_data + morton_offset;
+            data_ptrs[morton_to_linear ? 0 : 1] = &linear_data[linear_pixel_index];
+
+            std::memcpy(data_ptrs[0], data_ptrs[1], bytes_per_pixel);
+        }
+    }
+}
+
+} // namespace VideoCore

src/video_core/morton.h

@@ -0,0 +1,21 @@
+// Copyright 2018 yuzu Emulator Project
+// Licensed under GPLv2 or any later version
+// Refer to the license.txt file included.
+
+#pragma once
+
+#include "common/common_types.h"
+#include "video_core/surface.h"
+
+namespace VideoCore {
+
+enum class MortonSwizzleMode { MortonToLinear, LinearToMorton };
+
+void MortonSwizzle(MortonSwizzleMode mode, VideoCore::Surface::PixelFormat format, u32 stride,
+                   u32 block_height, u32 height, u32 block_depth, u32 depth, u8* buffer,
+                   std::size_t buffer_size, VAddr addr);
+
+void MortonCopyPixels128(u32 width, u32 height, u32 bytes_per_pixel, u32 linear_bytes_per_pixel,
+                         u8* morton_data, u8* linear_data, bool morton_to_linear);
+
+} // namespace VideoCore

src/video_core/renderer_opengl/gl_buffer_cache.cpp

@@ -76,7 +76,7 @@ std::tuple<u8*, GLintptr> OGLBufferCache::ReserveMemory(std::size_t size, std::s
     return std::make_tuple(uploaded_ptr, uploaded_offset);
 }
 
-void OGLBufferCache::Map(std::size_t max_size) {
+bool OGLBufferCache::Map(std::size_t max_size) {
     bool invalidate;
     std::tie(buffer_ptr, buffer_offset_base, invalidate) =
         stream_buffer.Map(static_cast<GLsizeiptr>(max_size), 4);
@@ -85,6 +85,7 @@ void OGLBufferCache::Map(std::size_t max_size) {
     if (invalidate) {
         InvalidateAll();
     }
+    return invalidate;
 }
 
 void OGLBufferCache::Unmap() {

src/video_core/renderer_opengl/gl_buffer_cache.h

@@ -50,7 +50,7 @@ public:
     /// Reserves memory to be used by host's CPU. Returns mapped address and offset.
     std::tuple<u8*, GLintptr> ReserveMemory(std::size_t size, std::size_t alignment = 4);
 
-    void Map(std::size_t max_size);
+    bool Map(std::size_t max_size);
     void Unmap();
 
     GLuint GetHandle() const;

src/video_core/renderer_opengl/gl_rasterizer.cpp

@@ -98,14 +98,9 @@ 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
@@ -181,15 +176,25 @@ void RasterizerOpenGL::SetupVertexFormat() {
     }
     state.draw.vertex_array = VAO.handle;
     state.ApplyVertexBufferState();
+
+    // Rebinding the VAO invalidates the vertex buffer bindings.
+    gpu.dirty_flags.vertex_array = 0xFFFFFFFF;
 }
 
 void RasterizerOpenGL::SetupVertexBuffer() {
-    MICROPROFILE_SCOPE(OpenGL_VB);
-    const auto& gpu = Core::System::GetInstance().GPU().Maxwell3D();
+    auto& gpu = Core::System::GetInstance().GPU().Maxwell3D();
     const auto& regs = gpu.regs;
 
+    if (!gpu.dirty_flags.vertex_array)
+        return;
+
+    MICROPROFILE_SCOPE(OpenGL_VB);
+
     // Upload all guest vertex arrays sequentially to our buffer
     for (u32 index = 0; index < Maxwell::NumVertexArrays; ++index) {
+        if (~gpu.dirty_flags.vertex_array & (1u << index))
+            continue;
+
         const auto& vertex_array = regs.vertex_array[index];
         if (!vertex_array.IsEnabled())
             continue;
@@ -216,6 +221,8 @@ void RasterizerOpenGL::SetupVertexBuffer() {
 
     // Implicit set by glBindVertexBuffer. Stupid glstate handling...
     state.draw.vertex_buffer = buffer_cache.GetHandle();
+
+    gpu.dirty_flags.vertex_array = 0;
 }
 
 DrawParameters RasterizerOpenGL::SetupDraw() {
@@ -542,6 +549,30 @@ 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) {
@@ -553,6 +584,14 @@ 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) {
@@ -573,7 +612,7 @@ void RasterizerOpenGL::DrawArrays() {
         return;
 
     MICROPROFILE_SCOPE(OpenGL_Drawing);
-    const auto& gpu = Core::System::GetInstance().GPU().Maxwell3D();
+    auto& gpu = Core::System::GetInstance().GPU().Maxwell3D();
     const auto& regs = gpu.regs;
 
     ScopeAcquireGLContext acquire_context{emu_window};
@@ -588,12 +627,12 @@ void RasterizerOpenGL::DrawArrays() {
     SyncLogicOpState();
     SyncCullMode();
     SyncPrimitiveRestart();
-    SyncScissorTest();
+    SyncScissorTest(state);
     // Alpha Testing is synced on shaders.
     SyncTransformFeedback();
     SyncPointState();
     CheckAlphaTests();
-
+    SyncPolygonOffset();
     // TODO(bunnei): Sync framebuffer_scale uniform here
     // TODO(bunnei): Sync scissorbox uniform(s) here
 
@@ -626,7 +665,11 @@ void RasterizerOpenGL::DrawArrays() {
     // Add space for at least 18 constant buffers
     buffer_size += Maxwell::MaxConstBuffers * (MaxConstbufferSize + uniform_buffer_alignment);
 
-    buffer_cache.Map(buffer_size);
+    bool invalidate = buffer_cache.Map(buffer_size);
+    if (invalidate) {
+        // As all cached buffers are invalidated, we need to recheck their state.
+        gpu.dirty_flags.vertex_attrib_format = 0xFFFFFFFF;
+    }
 
     SetupVertexFormat();
     SetupVertexBuffer();
@@ -815,7 +858,7 @@ void RasterizerOpenGL::SamplerInfo::SyncWithConfig(const Tegra::Texture::TSCEntr
     }
     const u32 bias = config.mip_lod_bias.Value();
     // Sign extend the 13-bit value.
-    const u32 mask = 1U << (13 - 1);
+    constexpr 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;
@@ -942,13 +985,25 @@ u32 RasterizerOpenGL::SetupTextures(Maxwell::ShaderStage stage, Shader& shader,
 
 void RasterizerOpenGL::SyncViewport(OpenGLState& current_state) {
     const auto& regs = Core::System::GetInstance().GPU().Maxwell3D().regs;
-    for (std::size_t i = 0; i < Tegra::Engines::Maxwell3D::Regs::NumViewports; i++) {
-        const MathUtil::Rectangle<s32> viewport_rect{regs.viewport_transform[i].GetRect()};
+    const bool geometry_shaders_enabled =
+        regs.IsShaderConfigEnabled(static_cast<size_t>(Maxwell::ShaderProgram::Geometry));
+    const std::size_t viewport_count =
+        geometry_shaders_enabled ? Tegra::Engines::Maxwell3D::Regs::NumViewports : 1;
+    for (std::size_t i = 0; i < viewport_count; i++) {
         auto& viewport = current_state.viewports[i];
-        viewport.x = viewport_rect.left;
-        viewport.y = viewport_rect.bottom;
-        viewport.width = static_cast<GLfloat>(viewport_rect.GetWidth());
-        viewport.height = static_cast<GLfloat>(viewport_rect.GetHeight());
+        const auto& src = regs.viewports[i];
+        if (regs.viewport_transform_enabled) {
+            const MathUtil::Rectangle<s32> viewport_rect{regs.viewport_transform[i].GetRect()};
+            viewport.x = viewport_rect.left;
+            viewport.y = viewport_rect.bottom;
+            viewport.width = viewport_rect.GetWidth();
+            viewport.height = viewport_rect.GetHeight();
+        } else {
+            viewport.x = src.x;
+            viewport.y = src.y;
+            viewport.width = src.width;
+            viewport.height = src.height;
+        }
         viewport.depth_range_far = regs.viewports[i].depth_range_far;
         viewport.depth_range_near = regs.viewports[i].depth_range_near;
     }
@@ -1120,11 +1175,15 @@ void RasterizerOpenGL::SyncLogicOpState() {
     state.logic_op.operation = MaxwellToGL::LogicOp(regs.logic_op.operation);
 }
 
-void RasterizerOpenGL::SyncScissorTest() {
+void RasterizerOpenGL::SyncScissorTest(OpenGLState& current_state) {
     const auto& regs = Core::System::GetInstance().GPU().Maxwell3D().regs;
-    for (std::size_t i = 0; i < Tegra::Engines::Maxwell3D::Regs::NumViewports; i++) {
+    const bool geometry_shaders_enabled =
+        regs.IsShaderConfigEnabled(static_cast<size_t>(Maxwell::ShaderProgram::Geometry));
+    const std::size_t viewport_count =
+        geometry_shaders_enabled ? Tegra::Engines::Maxwell3D::Regs::NumViewports : 1;
+    for (std::size_t i = 0; i < viewport_count; i++) {
         const auto& src = regs.scissor_test[i];
-        auto& dst = state.viewports[i].scissor;
+        auto& dst = current_state.viewports[i].scissor;
         dst.enabled = (src.enable != 0);
         if (dst.enabled == 0) {
             return;
@@ -1152,6 +1211,16 @@ 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

@@ -91,19 +91,20 @@ private:
         void SyncWithConfig(const Tegra::Texture::TSCEntry& info);
 
     private:
-        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;
+        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;
     };
 
     /**
@@ -171,7 +172,7 @@ private:
     void SyncMultiSampleState();
 
     /// Syncs the scissor test state to match the guest state
-    void SyncScissorTest();
+    void SyncScissorTest(OpenGLState& current_state);
 
     /// Syncs the transform feedback state to match the guest state
     void SyncTransformFeedback();
@@ -182,13 +183,14 @@ 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

@@ -15,6 +15,7 @@
 #include "core/memory.h"
 #include "core/settings.h"
 #include "video_core/engines/maxwell_3d.h"
+#include "video_core/morton.h"
 #include "video_core/renderer_opengl/gl_rasterizer.h"
 #include "video_core/renderer_opengl/gl_rasterizer_cache.h"
 #include "video_core/renderer_opengl/gl_state.h"
@@ -22,10 +23,11 @@
 #include "video_core/surface.h"
 #include "video_core/textures/astc.h"
 #include "video_core/textures/decoders.h"
-#include "video_core/utils.h"
 
 namespace OpenGL {
 
+using VideoCore::MortonSwizzle;
+using VideoCore::MortonSwizzleMode;
 using VideoCore::Surface::ComponentTypeFromDepthFormat;
 using VideoCore::Surface::ComponentTypeFromRenderTarget;
 using VideoCore::Surface::ComponentTypeFromTexture;
@@ -265,11 +267,11 @@ 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_ARB, GL_RGBA, GL_UNSIGNED_INT_8_8_8_8, ComponentType::UNorm,
+    {GL_COMPRESSED_RGBA_BPTC_UNORM, GL_RGBA, GL_UNSIGNED_INT_8_8_8_8, ComponentType::UNorm,
      true}, // BC7U
-    {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,
+    {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,
      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 +308,8 @@ 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_ARB, GL_RGBA, GL_UNSIGNED_INT_8_8_8_8,
-     ComponentType::UNorm, true},                                              // BC7U_SRGB
+    {GL_COMPRESSED_SRGB_ALPHA_BPTC_UNORM, 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 +348,7 @@ static GLenum SurfaceTargetToGL(SurfaceTarget target) {
     case SurfaceTarget::TextureCubemap:
         return GL_TEXTURE_CUBE_MAP;
     case SurfaceTarget::TextureCubeArray:
-        return GL_TEXTURE_CUBE_MAP_ARRAY_ARB;
+        return GL_TEXTURE_CUBE_MAP_ARRAY;
     }
     LOG_CRITICAL(Render_OpenGL, "Unimplemented texture target={}", static_cast<u32>(target));
     UNREACHABLE();
@@ -370,174 +372,7 @@ MathUtil::Rectangle<u32> SurfaceParams::GetRect(u32 mip_level) const {
     return {0, actual_height, MipWidth(mip_level), 0};
 }
 
-template <bool morton_to_gl, PixelFormat format>
-void MortonCopy(u32 stride, u32 block_height, u32 height, u32 block_depth, u32 depth, u8* gl_buffer,
-                std::size_t gl_buffer_size, VAddr addr) {
-    constexpr u32 bytes_per_pixel = GetBytesPerPixel(format);
-
-    // With the BCn formats (DXT and DXN), each 4x4 tile is swizzled instead of just individual
-    // pixel values.
-    const u32 tile_size_x{GetDefaultBlockWidth(format)};
-    const u32 tile_size_y{GetDefaultBlockHeight(format)};
-
-    if (morton_to_gl) {
-        Tegra::Texture::UnswizzleTexture(gl_buffer, addr, tile_size_x, tile_size_y, bytes_per_pixel,
-                                         stride, height, depth, block_height, block_depth);
-    } else {
-        Tegra::Texture::CopySwizzledData((stride + tile_size_x - 1) / tile_size_x,
-                                         (height + tile_size_y - 1) / tile_size_y, depth,
-                                         bytes_per_pixel, bytes_per_pixel, Memory::GetPointer(addr),
-                                         gl_buffer, false, block_height, block_depth);
-    }
-}
-
-using GLConversionArray = std::array<void (*)(u32, u32, u32, u32, u32, u8*, std::size_t, VAddr),
-                                     VideoCore::Surface::MaxPixelFormat>;
-
-static constexpr GLConversionArray morton_to_gl_fns = {
-    // clang-format off
-        MortonCopy<true, PixelFormat::ABGR8U>,
-        MortonCopy<true, PixelFormat::ABGR8S>,
-        MortonCopy<true, PixelFormat::ABGR8UI>,
-        MortonCopy<true, PixelFormat::B5G6R5U>,
-        MortonCopy<true, PixelFormat::A2B10G10R10U>,
-        MortonCopy<true, PixelFormat::A1B5G5R5U>,
-        MortonCopy<true, PixelFormat::R8U>,
-        MortonCopy<true, PixelFormat::R8UI>,
-        MortonCopy<true, PixelFormat::RGBA16F>,
-        MortonCopy<true, PixelFormat::RGBA16U>,
-        MortonCopy<true, PixelFormat::RGBA16UI>,
-        MortonCopy<true, PixelFormat::R11FG11FB10F>,
-        MortonCopy<true, PixelFormat::RGBA32UI>,
-        MortonCopy<true, PixelFormat::DXT1>,
-        MortonCopy<true, PixelFormat::DXT23>,
-        MortonCopy<true, PixelFormat::DXT45>,
-        MortonCopy<true, PixelFormat::DXN1>,
-        MortonCopy<true, PixelFormat::DXN2UNORM>,
-        MortonCopy<true, PixelFormat::DXN2SNORM>,
-        MortonCopy<true, PixelFormat::BC7U>,
-        MortonCopy<true, PixelFormat::BC6H_UF16>,
-        MortonCopy<true, PixelFormat::BC6H_SF16>,
-        MortonCopy<true, PixelFormat::ASTC_2D_4X4>,
-        MortonCopy<true, PixelFormat::G8R8U>,
-        MortonCopy<true, PixelFormat::G8R8S>,
-        MortonCopy<true, PixelFormat::BGRA8>,
-        MortonCopy<true, PixelFormat::RGBA32F>,
-        MortonCopy<true, PixelFormat::RG32F>,
-        MortonCopy<true, PixelFormat::R32F>,
-        MortonCopy<true, PixelFormat::R16F>,
-        MortonCopy<true, PixelFormat::R16U>,
-        MortonCopy<true, PixelFormat::R16S>,
-        MortonCopy<true, PixelFormat::R16UI>,
-        MortonCopy<true, PixelFormat::R16I>,
-        MortonCopy<true, PixelFormat::RG16>,
-        MortonCopy<true, PixelFormat::RG16F>,
-        MortonCopy<true, PixelFormat::RG16UI>,
-        MortonCopy<true, PixelFormat::RG16I>,
-        MortonCopy<true, PixelFormat::RG16S>,
-        MortonCopy<true, PixelFormat::RGB32F>,
-        MortonCopy<true, PixelFormat::RGBA8_SRGB>,
-        MortonCopy<true, PixelFormat::RG8U>,
-        MortonCopy<true, PixelFormat::RG8S>,
-        MortonCopy<true, PixelFormat::RG32UI>,
-        MortonCopy<true, PixelFormat::R32UI>,
-        MortonCopy<true, PixelFormat::ASTC_2D_8X8>,
-        MortonCopy<true, PixelFormat::ASTC_2D_8X5>,
-        MortonCopy<true, PixelFormat::ASTC_2D_5X4>,
-        MortonCopy<true, PixelFormat::BGRA8_SRGB>,
-        MortonCopy<true, PixelFormat::DXT1_SRGB>,
-        MortonCopy<true, PixelFormat::DXT23_SRGB>,
-        MortonCopy<true, PixelFormat::DXT45_SRGB>,
-        MortonCopy<true, PixelFormat::BC7U_SRGB>,
-        MortonCopy<true, PixelFormat::ASTC_2D_4X4_SRGB>,
-        MortonCopy<true, PixelFormat::ASTC_2D_8X8_SRGB>,
-        MortonCopy<true, PixelFormat::ASTC_2D_8X5_SRGB>,
-        MortonCopy<true, PixelFormat::ASTC_2D_5X4_SRGB>,
-        MortonCopy<true, PixelFormat::ASTC_2D_5X5>,
-        MortonCopy<true, PixelFormat::ASTC_2D_5X5_SRGB>,
-        MortonCopy<true, PixelFormat::ASTC_2D_10X8>,
-        MortonCopy<true, PixelFormat::ASTC_2D_10X8_SRGB>,
-        MortonCopy<true, PixelFormat::Z32F>,
-        MortonCopy<true, PixelFormat::Z16>,
-        MortonCopy<true, PixelFormat::Z24S8>,
-        MortonCopy<true, PixelFormat::S8Z24>,
-        MortonCopy<true, PixelFormat::Z32FS8>,
-    // clang-format on
-};
-
-static constexpr GLConversionArray gl_to_morton_fns = {
-    // clang-format off
-        MortonCopy<false, PixelFormat::ABGR8U>,
-        MortonCopy<false, PixelFormat::ABGR8S>,
-        MortonCopy<false, PixelFormat::ABGR8UI>,
-        MortonCopy<false, PixelFormat::B5G6R5U>,
-        MortonCopy<false, PixelFormat::A2B10G10R10U>,
-        MortonCopy<false, PixelFormat::A1B5G5R5U>,
-        MortonCopy<false, PixelFormat::R8U>,
-        MortonCopy<false, PixelFormat::R8UI>,
-        MortonCopy<false, PixelFormat::RGBA16F>,
-        MortonCopy<false, PixelFormat::RGBA16U>,
-        MortonCopy<false, PixelFormat::RGBA16UI>,
-        MortonCopy<false, PixelFormat::R11FG11FB10F>,
-        MortonCopy<false, PixelFormat::RGBA32UI>,
-        MortonCopy<false, PixelFormat::DXT1>,
-        MortonCopy<false, PixelFormat::DXT23>,
-        MortonCopy<false, PixelFormat::DXT45>,
-        MortonCopy<false, PixelFormat::DXN1>,
-        MortonCopy<false, PixelFormat::DXN2UNORM>,
-        MortonCopy<false, PixelFormat::DXN2SNORM>,
-        MortonCopy<false, PixelFormat::BC7U>,
-        MortonCopy<false, PixelFormat::BC6H_UF16>,
-        MortonCopy<false, PixelFormat::BC6H_SF16>,
-        // TODO(Subv): Swizzling ASTC formats are not supported
-        nullptr,
-        MortonCopy<false, PixelFormat::G8R8U>,
-        MortonCopy<false, PixelFormat::G8R8S>,
-        MortonCopy<false, PixelFormat::BGRA8>,
-        MortonCopy<false, PixelFormat::RGBA32F>,
-        MortonCopy<false, PixelFormat::RG32F>,
-        MortonCopy<false, PixelFormat::R32F>,
-        MortonCopy<false, PixelFormat::R16F>,
-        MortonCopy<false, PixelFormat::R16U>,
-        MortonCopy<false, PixelFormat::R16S>,
-        MortonCopy<false, PixelFormat::R16UI>,
-        MortonCopy<false, PixelFormat::R16I>,
-        MortonCopy<false, PixelFormat::RG16>,
-        MortonCopy<false, PixelFormat::RG16F>,
-        MortonCopy<false, PixelFormat::RG16UI>,
-        MortonCopy<false, PixelFormat::RG16I>,
-        MortonCopy<false, PixelFormat::RG16S>,
-        MortonCopy<false, PixelFormat::RGB32F>,
-        MortonCopy<false, PixelFormat::RGBA8_SRGB>,
-        MortonCopy<false, PixelFormat::RG8U>,
-        MortonCopy<false, PixelFormat::RG8S>,
-        MortonCopy<false, PixelFormat::RG32UI>,
-        MortonCopy<false, PixelFormat::R32UI>,
-        nullptr,
-        nullptr,
-        nullptr,
-        MortonCopy<false, PixelFormat::BGRA8_SRGB>,
-        MortonCopy<false, PixelFormat::DXT1_SRGB>,
-        MortonCopy<false, PixelFormat::DXT23_SRGB>,
-        MortonCopy<false, PixelFormat::DXT45_SRGB>,
-        MortonCopy<false, PixelFormat::BC7U_SRGB>,
-        nullptr,
-        nullptr,
-        nullptr,
-        nullptr,
-        nullptr,
-        nullptr,
-        nullptr,
-        nullptr,
-        MortonCopy<false, PixelFormat::Z32F>,
-        MortonCopy<false, PixelFormat::Z16>,
-        MortonCopy<false, PixelFormat::Z24S8>,
-        MortonCopy<false, PixelFormat::S8Z24>,
-        MortonCopy<false, PixelFormat::Z32FS8>,
-    // clang-format on
-};
-
-void SwizzleFunc(const GLConversionArray& functions, const SurfaceParams& params,
+void SwizzleFunc(const MortonSwizzleMode& mode, const SurfaceParams& params,
                  std::vector<u8>& gl_buffer, u32 mip_level) {
     u32 depth = params.MipDepth(mip_level);
     if (params.target == SurfaceTarget::Texture2D) {
@@ -550,19 +385,19 @@ void SwizzleFunc(const GLConversionArray& functions, const SurfaceParams& params
         const u64 layer_size = params.LayerMemorySize();
         const u64 gl_size = params.LayerSizeGL(mip_level);
         for (u32 i = 0; i < params.depth; i++) {
-            functions[static_cast<std::size_t>(params.pixel_format)](
-                params.MipWidth(mip_level), params.MipBlockHeight(mip_level),
-                params.MipHeight(mip_level), params.MipBlockDepth(mip_level), 1,
-                gl_buffer.data() + offset_gl, gl_size, params.addr + offset);
+            MortonSwizzle(mode, params.pixel_format, params.MipWidth(mip_level),
+                          params.MipBlockHeight(mip_level), params.MipHeight(mip_level),
+                          params.MipBlockDepth(mip_level), 1, gl_buffer.data() + offset_gl, gl_size,
+                          params.addr + offset);
             offset += layer_size;
             offset_gl += gl_size;
         }
     } else {
         const u64 offset = params.GetMipmapLevelOffset(mip_level);
-        functions[static_cast<std::size_t>(params.pixel_format)](
-            params.MipWidth(mip_level), params.MipBlockHeight(mip_level),
-            params.MipHeight(mip_level), params.MipBlockDepth(mip_level), depth, gl_buffer.data(),
-            gl_buffer.size(), params.addr + offset);
+        MortonSwizzle(mode, params.pixel_format, params.MipWidth(mip_level),
+                      params.MipBlockHeight(mip_level), params.MipHeight(mip_level),
+                      params.MipBlockDepth(mip_level), depth, gl_buffer.data(), gl_buffer.size(),
+                      params.addr + offset);
     }
 }
 
@@ -726,7 +561,7 @@ 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_ARB);
+    glBufferData(GL_PIXEL_PACK_BUFFER, buffer_size, nullptr, GL_STREAM_DRAW);
     if (source_format.compressed) {
         glGetCompressedTextureImage(src_surface->Texture().handle, src_attachment,
                                     static_cast<GLsizei>(src_params.size_in_bytes), nullptr);
@@ -903,18 +738,6 @@ static void ConvertS8Z24ToZ24S8(std::vector<u8>& data, u32 width, u32 height, bo
     }
 }
 
-static void ConvertG8R8ToR8G8(std::vector<u8>& data, u32 width, u32 height) {
-    constexpr auto bpp{GetBytesPerPixel(PixelFormat::G8R8U)};
-    for (std::size_t y = 0; y < height; ++y) {
-        for (std::size_t x = 0; x < width; ++x) {
-            const std::size_t offset{bpp * (y * width + x)};
-            const u8 temp{data[offset]};
-            data[offset] = data[offset + 1];
-            data[offset + 1] = temp;
-        }
-    }
-}
-
 /**
  * Helper function to perform software conversion (as needed) when loading a buffer from Switch
  * memory. This is for Maxwell pixel formats that cannot be represented as-is in OpenGL or with
@@ -947,12 +770,6 @@ static void ConvertFormatAsNeeded_LoadGLBuffer(std::vector<u8>& data, PixelForma
         // Convert the S8Z24 depth format to Z24S8, as OpenGL does not support S8Z24.
         ConvertS8Z24ToZ24S8(data, width, height, false);
         break;
-
-    case PixelFormat::G8R8U:
-    case PixelFormat::G8R8S:
-        // Convert the G8R8 color format to R8G8, as OpenGL does not support G8R8.
-        ConvertG8R8ToR8G8(data, width, height);
-        break;
     }
 }
 
@@ -996,7 +813,7 @@ void CachedSurface::LoadGLBuffer() {
         ASSERT_MSG(params.block_width == 1, "Block width is defined as {} on texture type {}",
                    params.block_width, static_cast<u32>(params.target));
         for (u32 i = 0; i < params.max_mip_level; i++)
-            SwizzleFunc(morton_to_gl_fns, params, gl_buffer[i], i);
+            SwizzleFunc(MortonSwizzleMode::MortonToLinear, params, gl_buffer[i], i);
     } else {
         const auto texture_src_data{Memory::GetPointer(params.addr)};
         const auto texture_src_data_end{texture_src_data + params.size_in_bytes_gl};
@@ -1035,7 +852,7 @@ void CachedSurface::FlushGLBuffer() {
         ASSERT_MSG(params.block_width == 1, "Block width is defined as {} on texture type {}",
                    params.block_width, static_cast<u32>(params.target));
 
-        SwizzleFunc(gl_to_morton_fns, params, gl_buffer[0], 0);
+        SwizzleFunc(MortonSwizzleMode::LinearToMorton, params, gl_buffer[0], 0);
     } else {
         std::memcpy(Memory::GetPointer(GetAddr()), gl_buffer[0].data(), GetSizeInBytes());
     }

src/video_core/renderer_opengl/gl_shader_cache.cpp

@@ -84,6 +84,7 @@ 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 GLShader::MAX_PROGRAM_CODE_LENGTH * sizeof(u64);
+        return shader_length;
     }
 
     // We do not have to flush this cache as things in it are never modified by us.
@@ -82,6 +82,7 @@ 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_gen.cpp

@@ -24,8 +24,7 @@ layout (location = 0) out vec4 position;
 
 layout(std140) uniform vs_config {
     vec4 viewport_flip;
-    uvec4 instance_id;
-    uvec4 flip_stage;
+    uvec4 config_pack; // instance_id, flip_stage, y_direction, padding
     uvec4 alpha_test;
 };
 )";
@@ -63,7 +62,8 @@ void main() {
     out += R"(
 
     // Check if the flip stage is VertexB
-    if (flip_stage[0] == 1) {
+    // Config pack's second value is flip_stage
+    if (config_pack[1] == 1) {
         // Viewport can be flipped, which is unsupported by glViewport
         position.xy *= viewport_flip.xy;
     }
@@ -71,7 +71,7 @@ void main() {
 
     // TODO(bunnei): This is likely a hack, position.w should be interpolated as 1.0
     // For now, this is here to bring order in lieu of proper emulation
-    if (flip_stage[0] == 1) {
+    if (config_pack[1] == 1) {
         position.w = 1.0;
     }
 }
@@ -101,8 +101,7 @@ layout (location = 0) out vec4 position;
 
 layout (std140) uniform gs_config {
     vec4 viewport_flip;
-    uvec4 instance_id;
-    uvec4 flip_stage;
+    uvec4 config_pack; // instance_id, flip_stage, y_direction, padding
     uvec4 alpha_test;
 };
 
@@ -139,8 +138,7 @@ layout (location = 0) in vec4 position;
 
 layout (std140) uniform fs_config {
     vec4 viewport_flip;
-    uvec4 instance_id;
-    uvec4 flip_stage;
+    uvec4 config_pack; // instance_id, flip_stage, y_direction, padding
     uvec4 alpha_test;
 };
 

src/video_core/renderer_opengl/gl_shader_gen.h

@@ -163,6 +163,7 @@ 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_shader_manager.cpp

@@ -27,16 +27,18 @@ void MaxwellUniformData::SetFromRegs(const Maxwell3D::State::ShaderStageInfo& sh
     alpha_test.func = func;
     alpha_test.ref = regs.alpha_test_ref;
 
-    // We only assign the instance to the first component of the vector, the rest is just padding.
-    instance_id[0] = state.current_instance;
+    instance_id = state.current_instance;
 
     // Assign in which stage the position has to be flipped
     // (the last stage before the fragment shader).
     if (gpu.regs.shader_config[static_cast<u32>(Maxwell3D::Regs::ShaderProgram::Geometry)].enable) {
-        flip_stage[0] = static_cast<u32>(Maxwell3D::Regs::ShaderProgram::Geometry);
+        flip_stage = static_cast<u32>(Maxwell3D::Regs::ShaderProgram::Geometry);
     } else {
-        flip_stage[0] = static_cast<u32>(Maxwell3D::Regs::ShaderProgram::VertexB);
+        flip_stage = static_cast<u32>(Maxwell3D::Regs::ShaderProgram::VertexB);
     }
+
+    // Y_NEGATE controls what value S2R returns for the Y_DIRECTION system value.
+    y_direction = regs.screen_y_control.y_negate == 0 ? 1.f : -1.f;
 }
 
 } // namespace OpenGL::GLShader

src/video_core/renderer_opengl/gl_shader_manager.h

@@ -21,8 +21,11 @@ using Tegra::Engines::Maxwell3D;
 struct MaxwellUniformData {
     void SetFromRegs(const Maxwell3D::State::ShaderStageInfo& shader_stage);
     alignas(16) GLvec4 viewport_flip;
-    alignas(16) GLuvec4 instance_id;
-    alignas(16) GLuvec4 flip_stage;
+    struct alignas(16) {
+        GLuint instance_id;
+        GLuint flip_stage;
+        GLfloat y_direction;
+    };
     struct alignas(16) {
         GLuint enabled;
         GLuint func;
@@ -30,7 +33,7 @@ struct MaxwellUniformData {
         GLuint padding;
     } alpha_test;
 };
-static_assert(sizeof(MaxwellUniformData) == 64, "MaxwellUniformData structure size is incorrect");
+static_assert(sizeof(MaxwellUniformData) == 48, "MaxwellUniformData structure size is incorrect");
 static_assert(sizeof(MaxwellUniformData) < 16384,
               "MaxwellUniformData structure must be less than 16kb as per the OpenGL spec");
 

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() {
@@ -233,6 +240,28 @@ 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) {
@@ -242,7 +271,9 @@ 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, updated.x, updated.y, updated.width, updated.height);
+                glViewportIndexedf(
+                    i, static_cast<GLfloat>(updated.x), static_cast<GLfloat>(updated.y),
+                    static_cast<GLfloat>(updated.width), static_cast<GLfloat>(updated.height));
             }
             if (updated.depth_range_near != current.depth_range_near ||
                 updated.depth_range_far != current.depth_range_far) {
@@ -270,8 +301,7 @@ 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(static_cast<GLint>(updated.x), static_cast<GLint>(updated.y),
-                       static_cast<GLsizei>(updated.width), static_cast<GLsizei>(updated.height));
+            glViewport(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) {
@@ -339,14 +369,14 @@ 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) {
-        glBlendFuncSeparateiARB(static_cast<GLuint>(target), updated.src_rgb_func,
-                                updated.dst_rgb_func, updated.src_a_func, updated.dst_a_func);
+        glBlendFuncSeparatei(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) {
-        glBlendEquationSeparateiARB(static_cast<GLuint>(target), updated.rgb_equation,
-                                    updated.a_equation);
+        glBlendEquationSeparatei(static_cast<GLuint>(target), updated.rgb_equation,
+                                 updated.a_equation);
     }
 }
 
@@ -383,6 +413,55 @@ 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];
@@ -509,6 +588,7 @@ 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 {
-        GLfloat x;
-        GLfloat y;
-        GLfloat width;
-        GLfloat height;
+        GLint x;
+        GLint y;
+        GLint width;
+        GLint 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();
@@ -206,6 +215,7 @@ public:
     OpenGLState& ResetBuffer(GLuint handle);
     OpenGLState& ResetVertexArray(GLuint handle);
     OpenGLState& ResetFramebuffer(GLuint handle);
+    void EmulateViewportWithScissor();
 
 private:
     static OpenGLState cur_state;
@@ -225,6 +235,7 @@ private:
     void ApplyLogicOp() const;
     void ApplyTextures() const;
     void ApplySamplers() const;
+    void ApplyPolygonOffset() const;
 };
 
 } // namespace OpenGL

src/video_core/renderer_opengl/renderer_opengl.cpp

@@ -19,9 +19,9 @@
 #include "core/settings.h"
 #include "core/telemetry_session.h"
 #include "core/tracer/recorder.h"
+#include "video_core/morton.h"
 #include "video_core/renderer_opengl/gl_rasterizer.h"
 #include "video_core/renderer_opengl/renderer_opengl.h"
-#include "video_core/utils.h"
 
 namespace OpenGL {
 
@@ -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_3_3) {
+    if (!GLAD_GL_VERSION_4_3) {
         return false;
     }
 

src/video_core/utils.h

@@ -1,164 +0,0 @@
-// Copyright 2014 Citra Emulator Project
-// Licensed under GPLv2 or any later version
-// Refer to the license.txt file included.
-
-#pragma once
-
-#include "common/common_types.h"
-
-namespace VideoCore {
-
-// 8x8 Z-Order coordinate from 2D coordinates
-static inline u32 MortonInterleave(u32 x, u32 y) {
-    static const u32 xlut[] = {0x00, 0x01, 0x04, 0x05, 0x10, 0x11, 0x14, 0x15};
-    static const u32 ylut[] = {0x00, 0x02, 0x08, 0x0a, 0x20, 0x22, 0x28, 0x2a};
-    return xlut[x % 8] + ylut[y % 8];
-}
-
-/**
- * Calculates the offset of the position of the pixel in Morton order
- */
-static inline u32 GetMortonOffset(u32 x, u32 y, u32 bytes_per_pixel) {
-    // Images are split into 8x8 tiles. Each tile is composed of four 4x4 subtiles each
-    // of which is composed of four 2x2 subtiles each of which is composed of four texels.
-    // Each structure is embedded into the next-bigger one in a diagonal pattern, e.g.
-    // texels are laid out in a 2x2 subtile like this:
-    // 2 3
-    // 0 1
-    //
-    // The full 8x8 tile has the texels arranged like this:
-    //
-    // 42 43 46 47 58 59 62 63
-    // 40 41 44 45 56 57 60 61
-    // 34 35 38 39 50 51 54 55
-    // 32 33 36 37 48 49 52 53
-    // 10 11 14 15 26 27 30 31
-    // 08 09 12 13 24 25 28 29
-    // 02 03 06 07 18 19 22 23
-    // 00 01 04 05 16 17 20 21
-    //
-    // This pattern is what's called Z-order curve, or Morton order.
-
-    const unsigned int block_height = 8;
-    const unsigned int coarse_x = x & ~7;
-
-    u32 i = VideoCore::MortonInterleave(x, y);
-
-    const unsigned int offset = coarse_x * block_height;
-
-    return (i + offset) * bytes_per_pixel;
-}
-
-static inline u32 MortonInterleave128(u32 x, u32 y) {
-    // 128x128 Z-Order coordinate from 2D coordinates
-    static constexpr u32 xlut[] = {
-        0x0000, 0x0001, 0x0002, 0x0003, 0x0008, 0x0009, 0x000a, 0x000b, 0x0040, 0x0041, 0x0042,
-        0x0043, 0x0048, 0x0049, 0x004a, 0x004b, 0x0800, 0x0801, 0x0802, 0x0803, 0x0808, 0x0809,
-        0x080a, 0x080b, 0x0840, 0x0841, 0x0842, 0x0843, 0x0848, 0x0849, 0x084a, 0x084b, 0x1000,
-        0x1001, 0x1002, 0x1003, 0x1008, 0x1009, 0x100a, 0x100b, 0x1040, 0x1041, 0x1042, 0x1043,
-        0x1048, 0x1049, 0x104a, 0x104b, 0x1800, 0x1801, 0x1802, 0x1803, 0x1808, 0x1809, 0x180a,
-        0x180b, 0x1840, 0x1841, 0x1842, 0x1843, 0x1848, 0x1849, 0x184a, 0x184b, 0x2000, 0x2001,
-        0x2002, 0x2003, 0x2008, 0x2009, 0x200a, 0x200b, 0x2040, 0x2041, 0x2042, 0x2043, 0x2048,
-        0x2049, 0x204a, 0x204b, 0x2800, 0x2801, 0x2802, 0x2803, 0x2808, 0x2809, 0x280a, 0x280b,
-        0x2840, 0x2841, 0x2842, 0x2843, 0x2848, 0x2849, 0x284a, 0x284b, 0x3000, 0x3001, 0x3002,
-        0x3003, 0x3008, 0x3009, 0x300a, 0x300b, 0x3040, 0x3041, 0x3042, 0x3043, 0x3048, 0x3049,
-        0x304a, 0x304b, 0x3800, 0x3801, 0x3802, 0x3803, 0x3808, 0x3809, 0x380a, 0x380b, 0x3840,
-        0x3841, 0x3842, 0x3843, 0x3848, 0x3849, 0x384a, 0x384b, 0x0000, 0x0001, 0x0002, 0x0003,
-        0x0008, 0x0009, 0x000a, 0x000b, 0x0040, 0x0041, 0x0042, 0x0043, 0x0048, 0x0049, 0x004a,
-        0x004b, 0x0800, 0x0801, 0x0802, 0x0803, 0x0808, 0x0809, 0x080a, 0x080b, 0x0840, 0x0841,
-        0x0842, 0x0843, 0x0848, 0x0849, 0x084a, 0x084b, 0x1000, 0x1001, 0x1002, 0x1003, 0x1008,
-        0x1009, 0x100a, 0x100b, 0x1040, 0x1041, 0x1042, 0x1043, 0x1048, 0x1049, 0x104a, 0x104b,
-        0x1800, 0x1801, 0x1802, 0x1803, 0x1808, 0x1809, 0x180a, 0x180b, 0x1840, 0x1841, 0x1842,
-        0x1843, 0x1848, 0x1849, 0x184a, 0x184b, 0x2000, 0x2001, 0x2002, 0x2003, 0x2008, 0x2009,
-        0x200a, 0x200b, 0x2040, 0x2041, 0x2042, 0x2043, 0x2048, 0x2049, 0x204a, 0x204b, 0x2800,
-        0x2801, 0x2802, 0x2803, 0x2808, 0x2809, 0x280a, 0x280b, 0x2840, 0x2841, 0x2842, 0x2843,
-        0x2848, 0x2849, 0x284a, 0x284b, 0x3000, 0x3001, 0x3002, 0x3003, 0x3008, 0x3009, 0x300a,
-        0x300b, 0x3040, 0x3041, 0x3042, 0x3043, 0x3048, 0x3049, 0x304a, 0x304b, 0x3800, 0x3801,
-        0x3802, 0x3803, 0x3808, 0x3809, 0x380a, 0x380b, 0x3840, 0x3841, 0x3842, 0x3843, 0x3848,
-        0x3849, 0x384a, 0x384b, 0x0000, 0x0001, 0x0002, 0x0003, 0x0008, 0x0009, 0x000a, 0x000b,
-        0x0040, 0x0041, 0x0042, 0x0043, 0x0048, 0x0049, 0x004a, 0x004b, 0x0800, 0x0801, 0x0802,
-        0x0803, 0x0808, 0x0809, 0x080a, 0x080b, 0x0840, 0x0841, 0x0842, 0x0843, 0x0848, 0x0849,
-        0x084a, 0x084b, 0x1000, 0x1001, 0x1002, 0x1003, 0x1008, 0x1009, 0x100a, 0x100b, 0x1040,
-        0x1041, 0x1042, 0x1043, 0x1048, 0x1049, 0x104a, 0x104b, 0x1800, 0x1801, 0x1802, 0x1803,
-        0x1808, 0x1809, 0x180a, 0x180b, 0x1840, 0x1841, 0x1842, 0x1843, 0x1848, 0x1849, 0x184a,
-        0x184b, 0x2000, 0x2001, 0x2002, 0x2003, 0x2008, 0x2009, 0x200a, 0x200b, 0x2040, 0x2041,
-        0x2042, 0x2043, 0x2048, 0x2049, 0x204a, 0x204b, 0x2800, 0x2801, 0x2802, 0x2803, 0x2808,
-        0x2809, 0x280a, 0x280b, 0x2840, 0x2841, 0x2842, 0x2843, 0x2848, 0x2849, 0x284a, 0x284b,
-        0x3000, 0x3001, 0x3002, 0x3003, 0x3008, 0x3009, 0x300a, 0x300b, 0x3040, 0x3041, 0x3042,
-        0x3043, 0x3048, 0x3049, 0x304a, 0x304b, 0x3800, 0x3801, 0x3802, 0x3803, 0x3808, 0x3809,
-        0x380a, 0x380b, 0x3840, 0x3841, 0x3842, 0x3843, 0x3848, 0x3849, 0x384a, 0x384b,
-    };
-    static constexpr u32 ylut[] = {
-        0x0000, 0x0004, 0x0010, 0x0014, 0x0020, 0x0024, 0x0030, 0x0034, 0x0080, 0x0084, 0x0090,
-        0x0094, 0x00a0, 0x00a4, 0x00b0, 0x00b4, 0x0100, 0x0104, 0x0110, 0x0114, 0x0120, 0x0124,
-        0x0130, 0x0134, 0x0180, 0x0184, 0x0190, 0x0194, 0x01a0, 0x01a4, 0x01b0, 0x01b4, 0x0200,
-        0x0204, 0x0210, 0x0214, 0x0220, 0x0224, 0x0230, 0x0234, 0x0280, 0x0284, 0x0290, 0x0294,
-        0x02a0, 0x02a4, 0x02b0, 0x02b4, 0x0300, 0x0304, 0x0310, 0x0314, 0x0320, 0x0324, 0x0330,
-        0x0334, 0x0380, 0x0384, 0x0390, 0x0394, 0x03a0, 0x03a4, 0x03b0, 0x03b4, 0x0400, 0x0404,
-        0x0410, 0x0414, 0x0420, 0x0424, 0x0430, 0x0434, 0x0480, 0x0484, 0x0490, 0x0494, 0x04a0,
-        0x04a4, 0x04b0, 0x04b4, 0x0500, 0x0504, 0x0510, 0x0514, 0x0520, 0x0524, 0x0530, 0x0534,
-        0x0580, 0x0584, 0x0590, 0x0594, 0x05a0, 0x05a4, 0x05b0, 0x05b4, 0x0600, 0x0604, 0x0610,
-        0x0614, 0x0620, 0x0624, 0x0630, 0x0634, 0x0680, 0x0684, 0x0690, 0x0694, 0x06a0, 0x06a4,
-        0x06b0, 0x06b4, 0x0700, 0x0704, 0x0710, 0x0714, 0x0720, 0x0724, 0x0730, 0x0734, 0x0780,
-        0x0784, 0x0790, 0x0794, 0x07a0, 0x07a4, 0x07b0, 0x07b4, 0x0000, 0x0004, 0x0010, 0x0014,
-        0x0020, 0x0024, 0x0030, 0x0034, 0x0080, 0x0084, 0x0090, 0x0094, 0x00a0, 0x00a4, 0x00b0,
-        0x00b4, 0x0100, 0x0104, 0x0110, 0x0114, 0x0120, 0x0124, 0x0130, 0x0134, 0x0180, 0x0184,
-        0x0190, 0x0194, 0x01a0, 0x01a4, 0x01b0, 0x01b4, 0x0200, 0x0204, 0x0210, 0x0214, 0x0220,
-        0x0224, 0x0230, 0x0234, 0x0280, 0x0284, 0x0290, 0x0294, 0x02a0, 0x02a4, 0x02b0, 0x02b4,
-        0x0300, 0x0304, 0x0310, 0x0314, 0x0320, 0x0324, 0x0330, 0x0334, 0x0380, 0x0384, 0x0390,
-        0x0394, 0x03a0, 0x03a4, 0x03b0, 0x03b4, 0x0400, 0x0404, 0x0410, 0x0414, 0x0420, 0x0424,
-        0x0430, 0x0434, 0x0480, 0x0484, 0x0490, 0x0494, 0x04a0, 0x04a4, 0x04b0, 0x04b4, 0x0500,
-        0x0504, 0x0510, 0x0514, 0x0520, 0x0524, 0x0530, 0x0534, 0x0580, 0x0584, 0x0590, 0x0594,
-        0x05a0, 0x05a4, 0x05b0, 0x05b4, 0x0600, 0x0604, 0x0610, 0x0614, 0x0620, 0x0624, 0x0630,
-        0x0634, 0x0680, 0x0684, 0x0690, 0x0694, 0x06a0, 0x06a4, 0x06b0, 0x06b4, 0x0700, 0x0704,
-        0x0710, 0x0714, 0x0720, 0x0724, 0x0730, 0x0734, 0x0780, 0x0784, 0x0790, 0x0794, 0x07a0,
-        0x07a4, 0x07b0, 0x07b4, 0x0000, 0x0004, 0x0010, 0x0014, 0x0020, 0x0024, 0x0030, 0x0034,
-        0x0080, 0x0084, 0x0090, 0x0094, 0x00a0, 0x00a4, 0x00b0, 0x00b4, 0x0100, 0x0104, 0x0110,
-        0x0114, 0x0120, 0x0124, 0x0130, 0x0134, 0x0180, 0x0184, 0x0190, 0x0194, 0x01a0, 0x01a4,
-        0x01b0, 0x01b4, 0x0200, 0x0204, 0x0210, 0x0214, 0x0220, 0x0224, 0x0230, 0x0234, 0x0280,
-        0x0284, 0x0290, 0x0294, 0x02a0, 0x02a4, 0x02b0, 0x02b4, 0x0300, 0x0304, 0x0310, 0x0314,
-        0x0320, 0x0324, 0x0330, 0x0334, 0x0380, 0x0384, 0x0390, 0x0394, 0x03a0, 0x03a4, 0x03b0,
-        0x03b4, 0x0400, 0x0404, 0x0410, 0x0414, 0x0420, 0x0424, 0x0430, 0x0434, 0x0480, 0x0484,
-        0x0490, 0x0494, 0x04a0, 0x04a4, 0x04b0, 0x04b4, 0x0500, 0x0504, 0x0510, 0x0514, 0x0520,
-        0x0524, 0x0530, 0x0534, 0x0580, 0x0584, 0x0590, 0x0594, 0x05a0, 0x05a4, 0x05b0, 0x05b4,
-        0x0600, 0x0604, 0x0610, 0x0614, 0x0620, 0x0624, 0x0630, 0x0634, 0x0680, 0x0684, 0x0690,
-        0x0694, 0x06a0, 0x06a4, 0x06b0, 0x06b4, 0x0700, 0x0704, 0x0710, 0x0714, 0x0720, 0x0724,
-        0x0730, 0x0734, 0x0780, 0x0784, 0x0790, 0x0794, 0x07a0, 0x07a4, 0x07b0, 0x07b4,
-    };
-    return xlut[x % 128] + ylut[y % 128];
-}
-
-static inline u32 GetMortonOffset128(u32 x, u32 y, u32 bytes_per_pixel) {
-    // Calculates the offset of the position of the pixel in Morton order
-    // Framebuffer images are split into 128x128 tiles.
-
-    const unsigned int block_height = 128;
-    const unsigned int coarse_x = x & ~127;
-
-    u32 i = MortonInterleave128(x, y);
-
-    const unsigned int offset = coarse_x * block_height;
-
-    return (i + offset) * bytes_per_pixel;
-}
-
-static inline void MortonCopyPixels128(u32 width, u32 height, u32 bytes_per_pixel,
-                                       u32 gl_bytes_per_pixel, u8* morton_data, u8* gl_data,
-                                       bool morton_to_gl) {
-    u8* data_ptrs[2];
-    for (unsigned y = 0; y < height; ++y) {
-        for (unsigned x = 0; x < width; ++x) {
-            const u32 coarse_y = y & ~127;
-            u32 morton_offset =
-                GetMortonOffset128(x, y, bytes_per_pixel) + coarse_y * width * bytes_per_pixel;
-            u32 gl_pixel_index = (x + y * width) * gl_bytes_per_pixel;
-
-            data_ptrs[morton_to_gl] = morton_data + morton_offset;
-            data_ptrs[!morton_to_gl] = &gl_data[gl_pixel_index];
-
-            memcpy(data_ptrs[0], data_ptrs[1], bytes_per_pixel);
-        }
-    }
-}
-
-} // namespace VideoCore

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::Submit);
-    connect(buttons, &QDialogButtonBox::rejected, this, &QtSoftwareKeyboardDialog::Reject);
+    connect(buttons, &QDialogButtonBox::accepted, this, &QtSoftwareKeyboardDialog::accept);
+    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::Submit() {
+void QtSoftwareKeyboardDialog::accept() {
     ok = true;
     text = line_edit->text().toStdU16String();
-    accept();
+    QDialog::accept();
 }
 
-void QtSoftwareKeyboardDialog::Reject() {
+void QtSoftwareKeyboardDialog::reject() {
     ok = false;
     text.clear();
-    accept();
+    QDialog::reject();
 }
 
 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 = out;
+    text_output = std::move(out);
     emit MainWindowGetText(parameters);
 }
 
 void QtSoftwareKeyboard::SendTextCheckDialog(std::u16string error_message,
                                              std::function<void()> finished_check) const {
-    this->finished_check = finished_check;
+    this->finished_check = std::move(finished_check);
     emit MainWindowTextCheckDialog(error_message);
 }
 

src/yuzu/applets/software_keyboard.h

@@ -33,8 +33,8 @@ public:
                              Core::Frontend::SoftwareKeyboardParameters parameters);
     ~QtSoftwareKeyboardDialog() override;
 
-    void Submit();
-    void Reject();
+    void accept() override;
+    void reject() override;
 
     std::u16string GetText() const;
     bool GetStatus() const;
@@ -70,11 +70,10 @@ signals:
     void MainWindowGetText(Core::Frontend::SoftwareKeyboardParameters parameters) const;
     void MainWindowTextCheckDialog(std::u16string error_message) const;
 
-public slots:
+private:
     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(3, 3);
+    fmt.setVersion(4, 3);
     fmt.setProfile(QGLFormat::CoreProfile);
     fmt.setSwapInterval(false);
 

src/yuzu/configuration/config.cpp

@@ -432,6 +432,7 @@ 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();
 
@@ -638,6 +639,7 @@ 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

@@ -34,6 +34,7 @@ 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);
 }
 
@@ -43,6 +44,7 @@ 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

@@ -145,6 +145,16 @@
         </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:(Currently does nothing.)</string>
+             <string>Internal Resolution</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

@@ -518,32 +518,18 @@ 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");
 
@@ -562,8 +548,8 @@ bool GMainWindow::LoadROM(const QString& filename) {
     render_window->MakeCurrent();
 
     if (!gladLoadGL()) {
-        QMessageBox::critical(this, tr("Error while initializing OpenGL 3.3 Core!"),
-                              tr("Your GPU may not support OpenGL 3.3, or you do not "
+        QMessageBox::critical(this, tr("Error while initializing OpenGL 4.3 Core!"),
+                              tr("Your GPU may not support OpenGL 4.3, or you do not "
                                  "have the latest graphics driver."));
         return false;
     }

src/yuzu_cmd/config.cpp

@@ -366,6 +366,7 @@ 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

@@ -206,6 +206,8 @@ 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

@@ -111,32 +111,18 @@ 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");
 
@@ -157,7 +143,7 @@ EmuWindow_SDL2::EmuWindow_SDL2(bool fullscreen) {
         exit(1);
     }
 
-    SDL_GL_SetAttribute(SDL_GL_CONTEXT_MAJOR_VERSION, 3);
+    SDL_GL_SetAttribute(SDL_GL_CONTEXT_MAJOR_VERSION, 4);
     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);