20 files changed, 1004 insertions, 329 deletions

diff --git a/src/core/arm/dyncom/arm_dyncom_interpreter.cpp b/src/core/arm/dyncom/arm_dyncom_interpreter.cpp
index 647784208..8d4b26815 100644
--- a/src/core/arm/dyncom/arm_dyncom_interpreter.cpp
+++ b/src/core/arm/dyncom/arm_dyncom_interpreter.cpp
@@ -10,7 +10,6 @@
 #include "common/common_types.h"
 #include "common/logging/log.h"
 #include "common/microprofile.h"
-#include "common/profiler.h"
 
 #include "core/memory.h"
 #include "core/hle/svc.h"
@@ -25,9 +24,6 @@
 
 #include "core/gdbstub/gdbstub.h"
 
-Common::Profiling::TimingCategory profile_execute("DynCom::Execute");
-Common::Profiling::TimingCategory profile_decode("DynCom::Decode");
-
 enum {
     COND            = (1 << 0),
     NON_BRANCH      = (1 << 1),
@@ -3496,7 +3492,6 @@ static unsigned int InterpreterTranslateInstruction(const ARMul_State* cpu, cons
 }
 
 static int InterpreterTranslateBlock(ARMul_State* cpu, int& bb_start, u32 addr) {
-    Common::Profiling::ScopeTimer timer_decode(profile_decode);
     MICROPROFILE_SCOPE(DynCom_Decode);
 
     // Decode instruction, get index
@@ -3530,7 +3525,6 @@ static int InterpreterTranslateBlock(ARMul_State* cpu, int& bb_start, u32 addr)
 }
 
 static int InterpreterTranslateSingle(ARMul_State* cpu, int& bb_start, u32 addr) {
-    Common::Profiling::ScopeTimer timer_decode(profile_decode);
     MICROPROFILE_SCOPE(DynCom_Decode);
 
     ARM_INST_PTR inst_base = nullptr;
@@ -3565,7 +3559,6 @@ static int clz(unsigned int x) {
 MICROPROFILE_DEFINE(DynCom_Execute, "DynCom", "Execute", MP_RGB(255, 0, 0));
 
 unsigned InterpreterMainLoop(ARMul_State* cpu) {
-    Common::Profiling::ScopeTimer timer_execute(profile_execute);
     MICROPROFILE_SCOPE(DynCom_Execute);
 
     GDBStub::BreakpointAddress breakpoint_data;
diff --git a/src/core/gdbstub/gdbstub.cpp b/src/core/gdbstub/gdbstub.cpp
index c1a7ec5bf..ae0c116ef 100644
--- a/src/core/gdbstub/gdbstub.cpp
+++ b/src/core/gdbstub/gdbstub.cpp
@@ -529,7 +529,7 @@ static void ReadRegister() {
         id |= HexCharToValue(command_buffer[2]);
     }
 
-    if (id >= R0_REGISTER && id <= R15_REGISTER) {
+    if (id <= R15_REGISTER) {
         IntToGdbHex(reply, Core::g_app_core->GetReg(id));
     } else if (id == CPSR_REGISTER) {
         IntToGdbHex(reply, Core::g_app_core->GetCPSR());
@@ -584,7 +584,7 @@ static void WriteRegister() {
         id |= HexCharToValue(command_buffer[2]);
     }
 
-    if (id >= R0_REGISTER && id <= R15_REGISTER) {
+    if (id <= R15_REGISTER) {
         Core::g_app_core->SetReg(id, GdbHexToInt(buffer_ptr));
     } else if (id == CPSR_REGISTER) {
         Core::g_app_core->SetCPSR(GdbHexToInt(buffer_ptr));
diff --git a/src/core/hle/result.h b/src/core/hle/result.h
index 2d22652d9..53931a106 100644
--- a/src/core/hle/result.h
+++ b/src/core/hle/result.h
@@ -18,6 +18,7 @@
 /// Detailed description of the error. This listing is likely incomplete.
 enum class ErrorDescription : u32 {
     Success = 0,
+    OS_InvalidBufferDescriptor = 48,
     WrongAddress = 53,
     FS_NotFound = 120,
     FS_AlreadyExists = 190,
diff --git a/src/core/hle/service/ac_u.cpp b/src/core/hle/service/ac_u.cpp
index d67325506..5241dd3e7 100644
--- a/src/core/hle/service/ac_u.cpp
+++ b/src/core/hle/service/ac_u.cpp
@@ -3,6 +3,8 @@
 // Refer to the license.txt file included.
 
 #include "common/logging/log.h"
+
+#include "core/hle/kernel/event.h"
 #include "core/hle/service/ac_u.h"
 
 ////////////////////////////////////////////////////////////////////////////////////////////////////
@@ -11,6 +13,28 @@
 namespace AC_U {
 
 /**
+ * AC_U::CloseAsync service function
+ *  Inputs:
+ *      1 : Always 0x20
+ *      3 : Always 0
+ *      4 : Event handle, should be signaled when AC connection is closed
+ *  Outputs:
+ *      1 : Result of function, 0 on success, otherwise error code
+ */
+static void CloseAsync(Service::Interface* self) {
+    u32* cmd_buff = Kernel::GetCommandBuffer();
+
+    auto evt = Kernel::g_handle_table.Get<Kernel::Event>(cmd_buff[4]);
+
+    if (evt) {
+        evt->name = "AC_U:close_event";
+        evt->Signal();
+    }
+    cmd_buff[1] = RESULT_SUCCESS.raw; // No error
+
+    LOG_WARNING(Service_AC, "(STUBBED) called");
+}
+/**
  * AC_U::GetWifiStatus service function
  *  Outputs:
  *      1 : Result of function, 0 on success, otherwise error code
@@ -47,7 +71,7 @@ const Interface::FunctionInfo FunctionTable[] = {
     {0x00010000, nullptr,               "CreateDefaultConfig"},
     {0x00040006, nullptr,               "ConnectAsync"},
     {0x00050002, nullptr,               "GetConnectResult"},
-    {0x00080004, nullptr,               "CloseAsync"},
+    {0x00080004, CloseAsync,            "CloseAsync"},
     {0x00090002, nullptr,               "GetCloseResult"},
     {0x000A0000, nullptr,               "GetLastErrorCode"},
     {0x000D0000, GetWifiStatus,         "GetWifiStatus"},
diff --git a/src/core/hle/service/am/am.cpp b/src/core/hle/service/am/am.cpp
index 9591522e5..3f71e7f2b 100644
--- a/src/core/hle/service/am/am.cpp
+++ b/src/core/hle/service/am/am.cpp
@@ -43,7 +43,7 @@ void FindContentInfos(Service::Interface* self) {
     am_content_count[media_type] = cmd_buff[4];
 
     cmd_buff[1] = RESULT_SUCCESS.raw;
-    LOG_WARNING(Service_AM, "(STUBBED) media_type=%u, title_id=0x%016lx, content_cound=%u, content_ids_pointer=0x%08x, content_info_pointer=0x%08x",
+    LOG_WARNING(Service_AM, "(STUBBED) media_type=%u, title_id=0x%016llx, content_cound=%u, content_ids_pointer=0x%08x, content_info_pointer=0x%08x",
                 media_type, title_id, am_content_count[media_type], content_ids_pointer, content_info_pointer);
 }
 
diff --git a/src/core/hle/service/dsp_dsp.cpp b/src/core/hle/service/dsp_dsp.cpp
index 08e437125..995bee3f9 100644
--- a/src/core/hle/service/dsp_dsp.cpp
+++ b/src/core/hle/service/dsp_dsp.cpp
@@ -2,6 +2,7 @@
 // Licensed under GPLv2 or any later version
 // Refer to the license.txt file included.
 
+#include <algorithm>
 #include <cinttypes>
 
 #include "audio_core/hle/pipe.h"
@@ -12,37 +13,80 @@
 #include "core/hle/kernel/event.h"
 #include "core/hle/service/dsp_dsp.h"
 
+using DspPipe = DSP::HLE::DspPipe;
+
 ////////////////////////////////////////////////////////////////////////////////////////////////////
 // Namespace DSP_DSP
 
 namespace DSP_DSP {
 
-static u32 read_pipe_count;
 static Kernel::SharedPtr<Kernel::Event> semaphore_event;
 
-struct PairHash {
-    template <typename T, typename U>
-    std::size_t operator()(const std::pair<T, U> &x) const {
-        // TODO(yuriks): Replace with better hash combining function.
-        return std::hash<T>()(x.first) ^ std::hash<U>()(x.second);
+/// There are three types of interrupts
+enum class InterruptType {
+    Zero, One, Pipe
+};
+constexpr size_t NUM_INTERRUPT_TYPE = 3;
+
+class InterruptEvents final {
+public:
+    void Signal(InterruptType type, DspPipe pipe) {
+        Kernel::SharedPtr<Kernel::Event>& event = Get(type, pipe);
+        if (event) {
+            event->Signal();
+        }
     }
+
+    Kernel::SharedPtr<Kernel::Event>& Get(InterruptType type, DspPipe dsp_pipe) {
+        switch (type) {
+        case InterruptType::Zero:
+            return zero;
+        case InterruptType::One:
+            return one;
+        case InterruptType::Pipe: {
+            const size_t pipe_index = static_cast<size_t>(dsp_pipe);
+            ASSERT(pipe_index < DSP::HLE::NUM_DSP_PIPE);
+            return pipe[pipe_index];
+        }
+        }
+
+        UNREACHABLE_MSG("Invalid interrupt type = %zu", static_cast<size_t>(type));
+    }
+
+    bool HasTooManyEventsRegistered() const {
+        // Actual service implementation only has 6 'slots' for interrupts.
+        constexpr size_t max_number_of_interrupt_events = 6;
+
+        size_t number = std::count_if(pipe.begin(), pipe.end(), [](const auto& evt) {
+            return evt != nullptr;
+        });
+
+        if (zero != nullptr)
+            number++;
+        if (one != nullptr)
+            number++;
+
+        return number >= max_number_of_interrupt_events;
+    }
+
+private:
+    /// Currently unknown purpose
+    Kernel::SharedPtr<Kernel::Event> zero = nullptr;
+    /// Currently unknown purpose
+    Kernel::SharedPtr<Kernel::Event> one = nullptr;
+    /// Each DSP pipe has an associated interrupt
+    std::array<Kernel::SharedPtr<Kernel::Event>, DSP::HLE::NUM_DSP_PIPE> pipe = {{}};
 };
 
-/// Map of (audio interrupt number, channel number) to Kernel::Events. See: RegisterInterruptEvents
-static std::unordered_map<std::pair<u32, u32>, Kernel::SharedPtr<Kernel::Event>, PairHash> interrupt_events;
+static InterruptEvents interrupt_events;
 
 // DSP Interrupts:
-// Interrupt #2 occurs every frame tick. Userland programs normally have a thread that's waiting
-// for an interrupt event. Immediately after this interrupt event, userland normally updates the
-// state in the next region and increments the relevant frame counter by two.
-void SignalAllInterrupts() {
-    // HACK: The other interrupts have currently unknown purpose, we trigger them each tick in any case.
-    for (auto& interrupt_event : interrupt_events)
-        interrupt_event.second->Signal();
-}
-
-void SignalInterrupt(u32 interrupt, u32 channel) {
-    interrupt_events[std::make_pair(interrupt, channel)]->Signal();
+// The audio-pipe interrupt occurs every frame tick. Userland programs normally have a thread
+// that's waiting for an interrupt event. Immediately after this interrupt event, userland
+// normally updates the state in the next region and increments the relevant frame counter by
+// two.
+void SignalPipeInterrupt(DspPipe pipe) {
+    interrupt_events.Signal(InterruptType::Pipe, pipe);
 }
 
 /**
@@ -58,7 +102,10 @@ static void ConvertProcessAddressFromDspDram(Service::Interface* self) {
 
     u32 addr = cmd_buff[1];
 
+    cmd_buff[0] = IPC::MakeHeader(0xC, 2, 0);
     cmd_buff[1] = RESULT_SUCCESS.raw; // No error
+
+    // TODO(merry): There is a per-region offset missing in this calculation (that seems to be always zero).
     cmd_buff[2] = (addr << 1) + (Memory::DSP_RAM_VADDR + 0x40000);
 
     LOG_DEBUG(Service_DSP, "addr=0x%08X", addr);
@@ -113,7 +160,9 @@ static void LoadComponent(Service::Interface* self) {
 static void GetSemaphoreEventHandle(Service::Interface* self) {
     u32* cmd_buff = Kernel::GetCommandBuffer();
 
+    cmd_buff[0] = IPC::MakeHeader(0x16, 1, 2);
     cmd_buff[1] = RESULT_SUCCESS.raw; // No error
+    // cmd_buff[2] not set
     cmd_buff[3] = Kernel::g_handle_table.Create(semaphore_event).MoveFrom(); // Event handle
 
     LOG_WARNING(Service_DSP, "(STUBBED) called");
@@ -138,8 +187,7 @@ static void FlushDataCache(Service::Interface* self) {
     u32 size    = cmd_buff[2];
     u32 process = cmd_buff[4];
 
-    // TODO(purpasmart96): Verify return header on HW
-
+    cmd_buff[0] = IPC::MakeHeader(0x13, 1, 0);
     cmd_buff[1] = RESULT_SUCCESS.raw; // No error
 
     LOG_TRACE(Service_DSP, "called address=0x%08X, size=0x%X, process=0x%08X", address, size, process);
@@ -148,8 +196,8 @@ static void FlushDataCache(Service::Interface* self) {
 /**
  * DSP_DSP::RegisterInterruptEvents service function
  *  Inputs:
- *      1 : Interrupt Number
- *      2 : Channel Number
+ *      1 : Interrupt Type
+ *      2 : Pipe Number
  *      4 : Interrupt event handle
  *  Outputs:
  *      1 : Result of function, 0 on success, otherwise error code
@@ -157,23 +205,40 @@ static void FlushDataCache(Service::Interface* self) {
 static void RegisterInterruptEvents(Service::Interface* self) {
     u32* cmd_buff = Kernel::GetCommandBuffer();
 
-    u32 interrupt = cmd_buff[1];
-    u32 channel = cmd_buff[2];
+    u32 type_index = cmd_buff[1];
+    u32 pipe_index = cmd_buff[2];
     u32 event_handle = cmd_buff[4];
 
+    ASSERT_MSG(type_index < NUM_INTERRUPT_TYPE && pipe_index < DSP::HLE::NUM_DSP_PIPE,
+               "Invalid type or pipe: type = %u, pipe = %u", type_index, pipe_index);
+
+    InterruptType type = static_cast<InterruptType>(cmd_buff[1]);
+    DspPipe pipe = static_cast<DspPipe>(cmd_buff[2]);
+
+    cmd_buff[0] = IPC::MakeHeader(0x15, 1, 0);
+
     if (event_handle) {
         auto evt = Kernel::g_handle_table.Get<Kernel::Event>(cmd_buff[4]);
-        if (evt) {
-            interrupt_events[std::make_pair(interrupt, channel)] = evt;
-            cmd_buff[1] = RESULT_SUCCESS.raw;
-            LOG_INFO(Service_DSP, "Registered interrupt=%u, channel=%u, event_handle=0x%08X", interrupt, channel, event_handle);
-        } else {
-            LOG_CRITICAL(Service_DSP, "Invalid event handle! interrupt=%u, channel=%u, event_handle=0x%08X", interrupt, channel, event_handle);
-            ASSERT(false); // This should really be handled at a IPC translation layer.
+
+        if (!evt) {
+            LOG_INFO(Service_DSP, "Invalid event handle! type=%u, pipe=%u, event_handle=0x%08X", type_index, pipe_index, event_handle);
+            ASSERT(false); // TODO: This should really be handled at an IPC translation layer.
+        }
+
+        if (interrupt_events.HasTooManyEventsRegistered()) {
+            LOG_INFO(Service_DSP, "Ran out of space to register interrupts (Attempted to register type=%u, pipe=%u, event_handle=0x%08X)",
+                     type_index, pipe_index, event_handle);
+            cmd_buff[1] = ResultCode(ErrorDescription::InvalidResultValue, ErrorModule::DSP, ErrorSummary::OutOfResource, ErrorLevel::Status).raw;
+            return;
         }
+
+        interrupt_events.Get(type, pipe) = evt;
+        LOG_INFO(Service_DSP, "Registered type=%u, pipe=%u, event_handle=0x%08X", type_index, pipe_index, event_handle);
+        cmd_buff[1] = RESULT_SUCCESS.raw;
     } else {
-        interrupt_events.erase(std::make_pair(interrupt, channel));
-        LOG_INFO(Service_DSP, "Unregistered interrupt=%u, channel=%u, event_handle=0x%08X", interrupt, channel, event_handle);
+        interrupt_events.Get(type, pipe) = nullptr;
+        LOG_INFO(Service_DSP, "Unregistered interrupt=%u, channel=%u, event_handle=0x%08X", type_index, pipe_index, event_handle);
+        cmd_buff[1] = RESULT_SUCCESS.raw;
     }
 }
 
@@ -187,6 +252,7 @@ static void RegisterInterruptEvents(Service::Interface* self) {
 static void SetSemaphore(Service::Interface* self) {
     u32* cmd_buff = Kernel::GetCommandBuffer();
 
+    cmd_buff[0] = IPC::MakeHeader(0x7, 1, 0);
     cmd_buff[1] = RESULT_SUCCESS.raw; // No error
 
     LOG_WARNING(Service_DSP, "(STUBBED) called");
@@ -195,7 +261,7 @@ static void SetSemaphore(Service::Interface* self) {
 /**
  * DSP_DSP::WriteProcessPipe service function
  *  Inputs:
- *      1 : Channel
+ *      1 : Pipe Number
  *      2 : Size
  *      3 : (size << 14) | 0x402
  *      4 : Buffer
@@ -206,24 +272,32 @@ static void SetSemaphore(Service::Interface* self) {
 static void WriteProcessPipe(Service::Interface* self) {
     u32* cmd_buff = Kernel::GetCommandBuffer();
 
-    DSP::HLE::DspPipe pipe = static_cast<DSP::HLE::DspPipe>(cmd_buff[1]);
+    u32 pipe_index = cmd_buff[1];
     u32 size = cmd_buff[2];
     u32 buffer = cmd_buff[4];
 
-    ASSERT_MSG(IPC::StaticBufferDesc(size, 1) == cmd_buff[3], "IPC static buffer descriptor failed validation (0x%X). pipe=%u, size=0x%X, buffer=0x%08X", cmd_buff[3], pipe, size, buffer);
-    ASSERT_MSG(Memory::GetPointer(buffer) != nullptr, "Invalid Buffer: pipe=%u, size=0x%X, buffer=0x%08X", pipe, size, buffer);
+    DSP::HLE::DspPipe pipe = static_cast<DSP::HLE::DspPipe>(pipe_index);
 
-    std::vector<u8> message(size);
+    if (IPC::StaticBufferDesc(size, 1) != cmd_buff[3]) {
+        LOG_ERROR(Service_DSP, "IPC static buffer descriptor failed validation (0x%X). pipe=%u, size=0x%X, buffer=0x%08X", cmd_buff[3], pipe_index, size, buffer);
+        cmd_buff[0] = IPC::MakeHeader(0, 1, 0);
+        cmd_buff[1] = ResultCode(ErrorDescription::OS_InvalidBufferDescriptor, ErrorModule::OS, ErrorSummary::WrongArgument, ErrorLevel::Permanent).raw;
+        return;
+    }
+
+    ASSERT_MSG(Memory::GetPointer(buffer) != nullptr, "Invalid Buffer: pipe=%u, size=0x%X, buffer=0x%08X", pipe_index, size, buffer);
 
+    std::vector<u8> message(size);
     for (size_t i = 0; i < size; i++) {
         message[i] = Memory::Read8(buffer + i);
     }
 
     DSP::HLE::PipeWrite(pipe, message);
 
+    cmd_buff[0] = IPC::MakeHeader(0xD, 1, 0);
     cmd_buff[1] = RESULT_SUCCESS.raw; // No error
 
-    LOG_DEBUG(Service_DSP, "pipe=%u, size=0x%X, buffer=0x%08X", pipe, size, buffer);
+    LOG_DEBUG(Service_DSP, "pipe=%u, size=0x%X, buffer=0x%08X", pipe_index, size, buffer);
 }
 
 /**
@@ -243,13 +317,16 @@ static void WriteProcessPipe(Service::Interface* self) {
 static void ReadPipeIfPossible(Service::Interface* self) {
     u32* cmd_buff = Kernel::GetCommandBuffer();
 
-    DSP::HLE::DspPipe pipe = static_cast<DSP::HLE::DspPipe>(cmd_buff[1]);
+    u32 pipe_index = cmd_buff[1];
     u32 unknown = cmd_buff[2];
     u32 size = cmd_buff[3] & 0xFFFF; // Lower 16 bits are size
     VAddr addr = cmd_buff[0x41];
 
-    ASSERT_MSG(Memory::GetPointer(addr) != nullptr, "Invalid addr: pipe=0x%08X, unknown=0x%08X, size=0x%X, buffer=0x%08X", pipe, unknown, size, addr);
+    DSP::HLE::DspPipe pipe = static_cast<DSP::HLE::DspPipe>(pipe_index);
+
+    ASSERT_MSG(Memory::GetPointer(addr) != nullptr, "Invalid addr: pipe=%u, unknown=0x%08X, size=0x%X, buffer=0x%08X", pipe_index, unknown, size, addr);
 
+    cmd_buff[0] = IPC::MakeHeader(0x10, 1, 2);
     cmd_buff[1] = RESULT_SUCCESS.raw; // No error
     if (DSP::HLE::GetPipeReadableSize(pipe) >= size) {
         std::vector<u8> response = DSP::HLE::PipeRead(pipe, size);
@@ -260,8 +337,10 @@ static void ReadPipeIfPossible(Service::Interface* self) {
     } else {
         cmd_buff[2] = 0; // Return no data
     }
+    cmd_buff[3] = IPC::StaticBufferDesc(size, 0);
+    cmd_buff[4] = addr;
 
-    LOG_DEBUG(Service_DSP, "pipe=0x%08X, unknown=0x%08X, size=0x%X, buffer=0x%08X, return cmd_buff[2]=0x%08X", pipe, unknown, size, addr, cmd_buff[2]);
+    LOG_DEBUG(Service_DSP, "pipe=%u, unknown=0x%08X, size=0x%X, buffer=0x%08X, return cmd_buff[2]=0x%08X", pipe_index, unknown, size, addr, cmd_buff[2]);
 }
 
 /**
@@ -278,26 +357,31 @@ static void ReadPipeIfPossible(Service::Interface* self) {
 static void ReadPipe(Service::Interface* self) {
     u32* cmd_buff = Kernel::GetCommandBuffer();
 
-    DSP::HLE::DspPipe pipe = static_cast<DSP::HLE::DspPipe>(cmd_buff[1]);
+    u32 pipe_index = cmd_buff[1];
     u32 unknown = cmd_buff[2];
     u32 size = cmd_buff[3] & 0xFFFF; // Lower 16 bits are size
     VAddr addr = cmd_buff[0x41];
 
-    ASSERT_MSG(Memory::GetPointer(addr) != nullptr, "Invalid addr: pipe=0x%08X, unknown=0x%08X, size=0x%X, buffer=0x%08X", pipe, unknown, size, addr);
+    DSP::HLE::DspPipe pipe = static_cast<DSP::HLE::DspPipe>(pipe_index);
+
+    ASSERT_MSG(Memory::GetPointer(addr) != nullptr, "Invalid addr: pipe=%u, unknown=0x%08X, size=0x%X, buffer=0x%08X", pipe_index, unknown, size, addr);
 
     if (DSP::HLE::GetPipeReadableSize(pipe) >= size) {
         std::vector<u8> response = DSP::HLE::PipeRead(pipe, size);
 
         Memory::WriteBlock(addr, response.data(), response.size());
 
+        cmd_buff[0] = IPC::MakeHeader(0xE, 2, 2);
         cmd_buff[1] = RESULT_SUCCESS.raw; // No error
         cmd_buff[2] = static_cast<u32>(response.size());
+        cmd_buff[3] = IPC::StaticBufferDesc(size, 0);
+        cmd_buff[4] = addr;
     } else {
         // No more data is in pipe. Hardware hangs in this case; this should never happen.
         UNREACHABLE();
     }
 
-    LOG_DEBUG(Service_DSP, "pipe=0x%08X, unknown=0x%08X, size=0x%X, buffer=0x%08X, return cmd_buff[2]=0x%08X", pipe, unknown, size, addr, cmd_buff[2]);
+    LOG_DEBUG(Service_DSP, "pipe=%u, unknown=0x%08X, size=0x%X, buffer=0x%08X, return cmd_buff[2]=0x%08X", pipe_index, unknown, size, addr, cmd_buff[2]);
 }
 
 /**
@@ -312,13 +396,16 @@ static void ReadPipe(Service::Interface* self) {
 static void GetPipeReadableSize(Service::Interface* self) {
     u32* cmd_buff = Kernel::GetCommandBuffer();
 
-    DSP::HLE::DspPipe pipe = static_cast<DSP::HLE::DspPipe>(cmd_buff[1]);
+    u32 pipe_index = cmd_buff[1];
     u32 unknown = cmd_buff[2];
 
+    DSP::HLE::DspPipe pipe = static_cast<DSP::HLE::DspPipe>(pipe_index);
+
+    cmd_buff[0] = IPC::MakeHeader(0xF, 2, 0);
     cmd_buff[1] = RESULT_SUCCESS.raw; // No error
     cmd_buff[2] = DSP::HLE::GetPipeReadableSize(pipe);
 
-    LOG_DEBUG(Service_DSP, "pipe=0x%08X, unknown=0x%08X, return cmd_buff[2]=0x%08X", pipe, unknown, cmd_buff[2]);
+    LOG_DEBUG(Service_DSP, "pipe=%u, unknown=0x%08X, return cmd_buff[2]=0x%08X", pipe_index, unknown, cmd_buff[2]);
 }
 
 /**
@@ -333,6 +420,7 @@ static void SetSemaphoreMask(Service::Interface* self) {
 
     u32 mask = cmd_buff[1];
 
+    cmd_buff[0] = IPC::MakeHeader(0x17, 1, 0);
     cmd_buff[1] = RESULT_SUCCESS.raw; // No error
 
     LOG_WARNING(Service_DSP, "(STUBBED) called mask=0x%08X", mask);
@@ -350,6 +438,7 @@ static void SetSemaphoreMask(Service::Interface* self) {
 static void GetHeadphoneStatus(Service::Interface* self) {
     u32* cmd_buff = Kernel::GetCommandBuffer();
 
+    cmd_buff[0] = IPC::MakeHeader(0x1F, 2, 0);
     cmd_buff[1] = RESULT_SUCCESS.raw; // No error
     cmd_buff[2] = 0; // Not using headphones?
 
@@ -376,6 +465,7 @@ static void RecvData(Service::Interface* self) {
 
     // Application reads this after requesting DSP shutdown, to verify the DSP has indeed shutdown or slept.
 
+    cmd_buff[0] = IPC::MakeHeader(0x1, 2, 0);
     cmd_buff[1] = RESULT_SUCCESS.raw;
     switch (DSP::HLE::GetDspState()) {
     case DSP::HLE::DspState::On:
@@ -411,6 +501,7 @@ static void RecvDataIsReady(Service::Interface* self) {
 
     ASSERT_MSG(register_number == 0, "Unknown register_number %u", register_number);
 
+    cmd_buff[0] = IPC::MakeHeader(0x2, 2, 0);
     cmd_buff[1] = RESULT_SUCCESS.raw;
     cmd_buff[2] = 1; // Ready to read
 
@@ -458,14 +549,14 @@ const Interface::FunctionInfo FunctionTable[] = {
 
 Interface::Interface() {
     semaphore_event = Kernel::Event::Create(Kernel::ResetType::OneShot, "DSP_DSP::semaphore_event");
-    read_pipe_count = 0;
+    interrupt_events = {};
 
     Register(FunctionTable);
 }
 
 Interface::~Interface() {
     semaphore_event = nullptr;
-    interrupt_events.clear();
+    interrupt_events = {};
 }
 
 } // namespace
diff --git a/src/core/hle/service/dsp_dsp.h b/src/core/hle/service/dsp_dsp.h
index 32b89e9bb..22f6687cc 100644
--- a/src/core/hle/service/dsp_dsp.h
+++ b/src/core/hle/service/dsp_dsp.h
@@ -8,6 +8,12 @@
 
 #include "core/hle/service/service.h"
 
+namespace DSP {
+namespace HLE {
+enum class DspPipe;
+}
+}
+
 ////////////////////////////////////////////////////////////////////////////////////////////////////
 // Namespace DSP_DSP
 
@@ -23,15 +29,10 @@ public:
     }
 };
 
-/// Signal all audio related interrupts.
-void SignalAllInterrupts();
-
 /**
- * Signal a specific audio related interrupt based on interrupt id and channel id.
- * @param interrupt_id The interrupt id
- * @param channel_id The channel id
- * The significance of various values of interrupt_id and channel_id is not yet known.
+ * Signal a specific DSP related interrupt of type == InterruptType::Pipe, pipe == pipe.
+ * @param pipe The DSP pipe for which to signal an interrupt for.
  */
-void SignalInterrupt(u32 interrupt_id, u32 channel_id);
+void SignalPipeInterrupt(DSP::HLE::DspPipe pipe);
 
-} // namespace
+} // namespace DSP_DSP
diff --git a/src/core/hle/service/fs/archive.cpp b/src/core/hle/service/fs/archive.cpp
index e9588cb72..cc51ede0c 100644
--- a/src/core/hle/service/fs/archive.cpp
+++ b/src/core/hle/service/fs/archive.cpp
@@ -114,6 +114,7 @@ ResultVal<bool> File::SyncRequest() {
                 return read.Code();
             }
             cmd_buff[2] = static_cast<u32>(*read);
+            Memory::RasterizerFlushAndInvalidateRegion(Memory::VirtualToPhysicalAddress(address), length);
             break;
         }
 
diff --git a/src/core/hle/service/fs/fs_user.cpp b/src/core/hle/service/fs/fs_user.cpp
index 3ec7ceb30..7df7da5a4 100644
--- a/src/core/hle/service/fs/fs_user.cpp
+++ b/src/core/hle/service/fs/fs_user.cpp
@@ -250,7 +250,7 @@ static void CreateFile(Service::Interface* self) {
 
     FileSys::Path file_path(filename_type, filename_size, filename_ptr);
 
-    LOG_DEBUG(Service_FS, "type=%d size=%llu data=%s", filename_type, filename_size, file_path.DebugStr().c_str());
+    LOG_DEBUG(Service_FS, "type=%d size=%llu data=%s", filename_type, file_size, file_path.DebugStr().c_str());
 
     cmd_buff[1] = CreateFileInArchive(archive_handle, file_path, file_size).raw;
 }
diff --git a/src/core/hle/service/gsp_gpu.cpp b/src/core/hle/service/gsp_gpu.cpp
index 0c655395e..b4c146e08 100644
--- a/src/core/hle/service/gsp_gpu.cpp
+++ b/src/core/hle/service/gsp_gpu.cpp
@@ -15,8 +15,6 @@
 
 #include "video_core/gpu_debugger.h"
 #include "video_core/debug_utils/debug_utils.h"
-#include "video_core/renderer_base.h"
-#include "video_core/video_core.h"
 
 #include "gsp_gpu.h"
 
@@ -45,6 +43,8 @@ Kernel::SharedPtr<Kernel::SharedMemory> g_shared_memory;
 /// Thread index into interrupt relay queue
 u32 g_thread_id = 0;
 
+static bool gpu_right_acquired = false;
+
 /// Gets a pointer to a thread command buffer in GSP shared memory
 static inline u8* GetCommandBuffer(u32 thread_id) {
     return g_shared_memory->GetPointer(0x800 + (thread_id * sizeof(CommandBuffer)));
@@ -291,8 +291,6 @@ static void FlushDataCache(Service::Interface* self) {
     u32 size    = cmd_buff[2];
     u32 process = cmd_buff[4];
 
-    VideoCore::g_renderer->Rasterizer()->InvalidateRegion(Memory::VirtualToPhysicalAddress(address), size);
-
     // TODO(purpasmart96): Verify return header on HW
 
     cmd_buff[1] = RESULT_SUCCESS.raw; // No error
@@ -374,6 +372,9 @@ static void UnregisterInterruptRelayQueue(Service::Interface* self) {
  * @todo This probably does not belong in the GSP module, instead move to video_core
  */
 void SignalInterrupt(InterruptId interrupt_id) {
+    if (!gpu_right_acquired) {
+        return;
+    }
     if (nullptr == g_interrupt_event) {
         LOG_WARNING(Service_GSP, "cannot synchronize until GSP event has been created!");
         return;
@@ -408,6 +409,8 @@ void SignalInterrupt(InterruptId interrupt_id) {
     g_interrupt_event->Signal();
 }
 
+MICROPROFILE_DEFINE(GPU_GSP_DMA, "GPU", "GSP DMA", MP_RGB(100, 0, 255));
+
 /// Executes the next GSP command
 static void ExecuteCommand(const Command& command, u32 thread_id) {
     // Utility function to convert register ID to address
@@ -419,18 +422,21 @@ static void ExecuteCommand(const Command& command, u32 thread_id) {
 
     // GX request DMA - typically used for copying memory from GSP heap to VRAM
     case CommandId::REQUEST_DMA:
-        VideoCore::g_renderer->Rasterizer()->FlushRegion(Memory::VirtualToPhysicalAddress(command.dma_request.source_address),
-                                                            command.dma_request.size);
+    {
+        MICROPROFILE_SCOPE(GPU_GSP_DMA);
+
+        // TODO: Consider attempting rasterizer-accelerated surface blit if that usage is ever possible/likely
+        Memory::RasterizerFlushRegion(Memory::VirtualToPhysicalAddress(command.dma_request.source_address),
+                            command.dma_request.size);
+        Memory::RasterizerFlushAndInvalidateRegion(Memory::VirtualToPhysicalAddress(command.dma_request.dest_address),
+                            command.dma_request.size);
 
         memcpy(Memory::GetPointer(command.dma_request.dest_address),
                Memory::GetPointer(command.dma_request.source_address),
                command.dma_request.size);
         SignalInterrupt(InterruptId::DMA);
-
-        VideoCore::g_renderer->Rasterizer()->InvalidateRegion(Memory::VirtualToPhysicalAddress(command.dma_request.dest_address),
-                                                          command.dma_request.size);
         break;
-
+    }
     // TODO: This will need some rework in the future. (why?)
     case CommandId::SUBMIT_GPU_CMDLIST:
     {
@@ -517,13 +523,8 @@ static void ExecuteCommand(const Command& command, u32 thread_id) {
 
     case CommandId::CACHE_FLUSH:
     {
-        for (auto& region : command.cache_flush.regions) {
-            if (region.size == 0)
-                break;
-
-            VideoCore::g_renderer->Rasterizer()->InvalidateRegion(
-                Memory::VirtualToPhysicalAddress(region.address), region.size);
-        }
+        // NOTE: Rasterizer flushing handled elsewhere in CPU read/write and other GPU handlers
+        // Use command.cache_flush.regions to implement this handler
         break;
     }
 
@@ -628,6 +629,35 @@ static void ImportDisplayCaptureInfo(Service::Interface* self) {
     LOG_WARNING(Service_GSP, "called");
 }
 
+/**
+ * GSP_GPU::AcquireRight service function
+ *  Outputs:
+ *      1: Result code
+ */
+static void AcquireRight(Service::Interface* self) {
+    u32* cmd_buff = Kernel::GetCommandBuffer();
+
+    gpu_right_acquired = true;
+
+    cmd_buff[1] = RESULT_SUCCESS.raw;
+
+    LOG_WARNING(Service_GSP, "called");
+}
+
+/**
+ * GSP_GPU::ReleaseRight service function
+ *  Outputs:
+ *      1: Result code
+ */
+static void ReleaseRight(Service::Interface* self) {
+    u32* cmd_buff = Kernel::GetCommandBuffer();
+
+    gpu_right_acquired = false;
+
+    cmd_buff[1] = RESULT_SUCCESS.raw;
+
+    LOG_WARNING(Service_GSP, "called");
+}
 
 const Interface::FunctionInfo FunctionTable[] = {
     {0x00010082, WriteHWRegs,                   "WriteHWRegs"},
@@ -651,8 +681,8 @@ const Interface::FunctionInfo FunctionTable[] = {
     {0x00130042, RegisterInterruptRelayQueue,   "RegisterInterruptRelayQueue"},
     {0x00140000, UnregisterInterruptRelayQueue, "UnregisterInterruptRelayQueue"},
     {0x00150002, nullptr,                       "TryAcquireRight"},
-    {0x00160042, nullptr,                       "AcquireRight"},
-    {0x00170000, nullptr,                       "ReleaseRight"},
+    {0x00160042, AcquireRight,                  "AcquireRight"},
+    {0x00170000, ReleaseRight,                  "ReleaseRight"},
     {0x00180000, ImportDisplayCaptureInfo,      "ImportDisplayCaptureInfo"},
     {0x00190000, nullptr,                       "SaveVramSysArea"},
     {0x001A0000, nullptr,                       "RestoreVramSysArea"},
@@ -673,11 +703,13 @@ Interface::Interface() {
     g_shared_memory = nullptr;
 
     g_thread_id = 0;
+    gpu_right_acquired = false;
 }
 
 Interface::~Interface() {
     g_interrupt_event = nullptr;
     g_shared_memory = nullptr;
+    gpu_right_acquired = false;
 }
 
 } // namespace
diff --git a/src/core/hle/service/y2r_u.cpp b/src/core/hle/service/y2r_u.cpp
index 22f373adf..d16578f87 100644
--- a/src/core/hle/service/y2r_u.cpp
+++ b/src/core/hle/service/y2r_u.cpp
@@ -4,6 +4,7 @@
 
 #include <cstring>
 
+#include "common/common_funcs.h"
 #include "common/common_types.h"
 #include "common/logging/log.h"
 
@@ -12,9 +13,6 @@
 #include "core/hle/service/y2r_u.h"
 #include "core/hw/y2r.h"
 
-#include "video_core/renderer_base.h"
-#include "video_core/video_core.h"
-
 ////////////////////////////////////////////////////////////////////////////////////////////////////
 // Namespace Y2R_U
 
@@ -28,13 +26,17 @@ struct ConversionParameters {
     u16 input_line_width;
     u16 input_lines;
     StandardCoefficient standard_coefficient;
-    u8 reserved;
+    u8 padding;
     u16 alpha;
 };
 static_assert(sizeof(ConversionParameters) == 12, "ConversionParameters struct has incorrect size");
 
 static Kernel::SharedPtr<Kernel::Event> completion_event;
 static ConversionConfiguration conversion;
+static DitheringWeightParams dithering_weight_params;
+static u32 temporal_dithering_enabled = 0;
+static u32 transfer_end_interrupt_enabled = 0;
+static u32 spacial_dithering_enabled = 0;
 
 static const CoefficientSet standard_coefficients[4] = {
     {{ 0x100, 0x166, 0xB6, 0x58, 0x1C5, -0x166F, 0x10EE, -0x1C5B }}, // ITU_Rec601
@@ -73,7 +75,7 @@ ResultCode ConversionConfiguration::SetInputLines(u16 lines) {
 
 ResultCode ConversionConfiguration::SetStandardCoefficient(StandardCoefficient standard_coefficient) {
     size_t index = static_cast<size_t>(standard_coefficient);
-    if (index >= 4) {
+    if (index >= ARRAY_SIZE(standard_coefficients)) {
         return ResultCode(ErrorDescription::InvalidEnumValue, ErrorModule::CAM,
             ErrorSummary::InvalidArgument, ErrorLevel::Usage); // 0xE0E053ED
     }
@@ -86,44 +88,183 @@ static void SetInputFormat(Service::Interface* self) {
     u32* cmd_buff = Kernel::GetCommandBuffer();
 
     conversion.input_format = static_cast<InputFormat>(cmd_buff[1]);
+
+    cmd_buff[0] = IPC::MakeHeader(0x1, 1, 0);
+    cmd_buff[1] = RESULT_SUCCESS.raw;
+
     LOG_DEBUG(Service_Y2R, "called input_format=%hhu", conversion.input_format);
+}
+
+static void GetInputFormat(Service::Interface* self) {
+    u32* cmd_buff = Kernel::GetCommandBuffer();
 
+    cmd_buff[0] = IPC::MakeHeader(0x2, 2, 0);
     cmd_buff[1] = RESULT_SUCCESS.raw;
+    cmd_buff[2] = static_cast<u32>(conversion.input_format);
+
+    LOG_DEBUG(Service_Y2R, "called input_format=%hhu", conversion.input_format);
 }
 
 static void SetOutputFormat(Service::Interface* self) {
     u32* cmd_buff = Kernel::GetCommandBuffer();
 
     conversion.output_format = static_cast<OutputFormat>(cmd_buff[1]);
+
+    cmd_buff[0] = IPC::MakeHeader(0x3, 1, 0);
+    cmd_buff[1] = RESULT_SUCCESS.raw;
+
     LOG_DEBUG(Service_Y2R, "called output_format=%hhu", conversion.output_format);
+}
+
+static void GetOutputFormat(Service::Interface* self) {
+    u32* cmd_buff = Kernel::GetCommandBuffer();
 
+    cmd_buff[0] = IPC::MakeHeader(0x4, 2, 0);
     cmd_buff[1] = RESULT_SUCCESS.raw;
+    cmd_buff[2] = static_cast<u32>(conversion.output_format);
+
+    LOG_DEBUG(Service_Y2R, "called output_format=%hhu", conversion.output_format);
 }
 
 static void SetRotation(Service::Interface* self) {
     u32* cmd_buff = Kernel::GetCommandBuffer();
 
     conversion.rotation = static_cast<Rotation>(cmd_buff[1]);
+
+    cmd_buff[0] = IPC::MakeHeader(0x5, 1, 0);
+    cmd_buff[1] = RESULT_SUCCESS.raw;
+
     LOG_DEBUG(Service_Y2R, "called rotation=%hhu", conversion.rotation);
+}
+
+static void GetRotation(Service::Interface* self) {
+    u32* cmd_buff = Kernel::GetCommandBuffer();
 
+    cmd_buff[0] = IPC::MakeHeader(0x6, 2, 0);
     cmd_buff[1] = RESULT_SUCCESS.raw;
+    cmd_buff[2] = static_cast<u32>(conversion.rotation);
+
+    LOG_DEBUG(Service_Y2R, "called rotation=%hhu", conversion.rotation);
 }
 
 static void SetBlockAlignment(Service::Interface* self) {
     u32* cmd_buff = Kernel::GetCommandBuffer();
 
     conversion.block_alignment = static_cast<BlockAlignment>(cmd_buff[1]);
-    LOG_DEBUG(Service_Y2R, "called alignment=%hhu", conversion.block_alignment);
 
+    cmd_buff[0] = IPC::MakeHeader(0x7, 1, 0);
+    cmd_buff[1] = RESULT_SUCCESS.raw;
+
+    LOG_DEBUG(Service_Y2R, "called block_alignment=%hhu", conversion.block_alignment);
+}
+
+static void GetBlockAlignment(Service::Interface* self) {
+    u32* cmd_buff = Kernel::GetCommandBuffer();
+
+    cmd_buff[0] = IPC::MakeHeader(0x8, 2, 0);
+    cmd_buff[1] = RESULT_SUCCESS.raw;
+    cmd_buff[2] = static_cast<u32>(conversion.block_alignment);
+
+    LOG_DEBUG(Service_Y2R, "called block_alignment=%hhu", conversion.block_alignment);
+}
+
+/**
+ * Y2R_U::SetSpacialDithering service function
+ *  Inputs:
+ *      1 : u8, 0 = Disabled, 1 = Enabled
+ *  Outputs:
+ *      1 : Result of function, 0 on success, otherwise error code
+ */
+static void SetSpacialDithering(Service::Interface* self) {
+    u32* cmd_buff = Kernel::GetCommandBuffer();
+    spacial_dithering_enabled = cmd_buff[1] & 0xF;
+
+    cmd_buff[0] = IPC::MakeHeader(0x9, 1, 0);
+    cmd_buff[1] = RESULT_SUCCESS.raw;
+
+    LOG_WARNING(Service_Y2R, "(STUBBED) called");
+}
+
+/**
+ * Y2R_U::GetSpacialDithering service function
+ *  Outputs:
+ *      1 : Result of function, 0 on success, otherwise error code
+ *      2 : u8, 0 = Disabled, 1 = Enabled
+ */
+static void GetSpacialDithering(Service::Interface* self) {
+    u32* cmd_buff = Kernel::GetCommandBuffer();
+
+    cmd_buff[0] = IPC::MakeHeader(0xA, 2, 0);
+    cmd_buff[1] = RESULT_SUCCESS.raw;
+    cmd_buff[2] = spacial_dithering_enabled;
+
+    LOG_WARNING(Service_Y2R, "(STUBBED) called");
+}
+
+/**
+ * Y2R_U::SetTemporalDithering service function
+ *  Inputs:
+ *      1 : u8, 0 = Disabled, 1 = Enabled
+ *  Outputs:
+ *      1 : Result of function, 0 on success, otherwise error code
+ */
+static void SetTemporalDithering(Service::Interface* self) {
+    u32* cmd_buff = Kernel::GetCommandBuffer();
+    temporal_dithering_enabled = cmd_buff[1] & 0xF;
+
+    cmd_buff[0] = IPC::MakeHeader(0xB, 1, 0);
     cmd_buff[1] = RESULT_SUCCESS.raw;
+
+    LOG_WARNING(Service_Y2R, "(STUBBED) called");
 }
 
+/**
+ * Y2R_U::GetTemporalDithering service function
+ *  Outputs:
+ *      1 : Result of function, 0 on success, otherwise error code
+ *      2 : u8, 0 = Disabled, 1 = Enabled
+ */
+static void GetTemporalDithering(Service::Interface* self) {
+    u32* cmd_buff = Kernel::GetCommandBuffer();
+
+    cmd_buff[0] = IPC::MakeHeader(0xC, 2, 0);
+    cmd_buff[1] = RESULT_SUCCESS.raw;
+    cmd_buff[2] = temporal_dithering_enabled;
+
+    LOG_WARNING(Service_Y2R, "(STUBBED) called");
+}
+
+/**
+ * Y2R_U::SetTransferEndInterrupt service function
+ *  Inputs:
+ *      1 : u8, 0 = Disabled, 1 = Enabled
+ *  Outputs:
+ *      1 : Result of function, 0 on success, otherwise error code
+ */
 static void SetTransferEndInterrupt(Service::Interface* self) {
     u32* cmd_buff = Kernel::GetCommandBuffer();
+    transfer_end_interrupt_enabled = cmd_buff[1] & 0xf;
 
     cmd_buff[0] = IPC::MakeHeader(0xD, 1, 0);
     cmd_buff[1] = RESULT_SUCCESS.raw;
-    LOG_DEBUG(Service_Y2R, "(STUBBED) called");
+
+    LOG_WARNING(Service_Y2R, "(STUBBED) called");
+}
+
+/**
+ * Y2R_U::GetTransferEndInterrupt service function
+ *  Outputs:
+ *      1 : Result of function, 0 on success, otherwise error code
+ *      2 : u8, 0 = Disabled, 1 = Enabled
+ */
+static void GetTransferEndInterrupt(Service::Interface* self) {
+    u32* cmd_buff = Kernel::GetCommandBuffer();
+
+    cmd_buff[0] = IPC::MakeHeader(0xE, 2, 0);
+    cmd_buff[1] = RESULT_SUCCESS.raw;
+    cmd_buff[2] = transfer_end_interrupt_enabled;
+
+    LOG_WARNING(Service_Y2R, "(STUBBED) called");
 }
 
 /**
@@ -135,8 +276,10 @@ static void SetTransferEndInterrupt(Service::Interface* self) {
 static void GetTransferEndEvent(Service::Interface* self) {
     u32* cmd_buff = Kernel::GetCommandBuffer();
 
+    cmd_buff[0] = IPC::MakeHeader(0xF, 2, 0);
     cmd_buff[1] = RESULT_SUCCESS.raw;
     cmd_buff[3] = Kernel::g_handle_table.Create(completion_event).MoveFrom();
+
     LOG_DEBUG(Service_Y2R, "called");
 }
 
@@ -147,12 +290,12 @@ static void SetSendingY(Service::Interface* self) {
     conversion.src_Y.image_size = cmd_buff[2];
     conversion.src_Y.transfer_unit = cmd_buff[3];
     conversion.src_Y.gap = cmd_buff[4];
-    u32 src_process_handle = cmd_buff[6];
-    LOG_DEBUG(Service_Y2R, "called image_size=0x%08X, transfer_unit=%hu, transfer_stride=%hu, "
-        "src_process_handle=0x%08X", conversion.src_Y.image_size,
-        conversion.src_Y.transfer_unit, conversion.src_Y.gap, src_process_handle);
 
+    cmd_buff[0] = IPC::MakeHeader(0x10, 1, 0);
     cmd_buff[1] = RESULT_SUCCESS.raw;
+
+    LOG_DEBUG(Service_Y2R, "called image_size=0x%08X, transfer_unit=%hu, transfer_stride=%hu, src_process_handle=0x%08X",
+            conversion.src_Y.image_size, conversion.src_Y.transfer_unit, conversion.src_Y.gap, cmd_buff[6]);
 }
 
 static void SetSendingU(Service::Interface* self) {
@@ -162,12 +305,12 @@ static void SetSendingU(Service::Interface* self) {
     conversion.src_U.image_size = cmd_buff[2];
     conversion.src_U.transfer_unit = cmd_buff[3];
     conversion.src_U.gap = cmd_buff[4];
-    u32 src_process_handle = cmd_buff[6];
-    LOG_DEBUG(Service_Y2R, "called image_size=0x%08X, transfer_unit=%hu, transfer_stride=%hu, "
-        "src_process_handle=0x%08X", conversion.src_U.image_size,
-        conversion.src_U.transfer_unit, conversion.src_U.gap, src_process_handle);
 
+    cmd_buff[0] = IPC::MakeHeader(0x11, 1, 0);
     cmd_buff[1] = RESULT_SUCCESS.raw;
+
+    LOG_DEBUG(Service_Y2R, "called image_size=0x%08X, transfer_unit=%hu, transfer_stride=%hu, src_process_handle=0x%08X",
+            conversion.src_U.image_size, conversion.src_U.transfer_unit, conversion.src_U.gap, cmd_buff[6]);
 }
 
 static void SetSendingV(Service::Interface* self) {
@@ -177,12 +320,12 @@ static void SetSendingV(Service::Interface* self) {
     conversion.src_V.image_size = cmd_buff[2];
     conversion.src_V.transfer_unit = cmd_buff[3];
     conversion.src_V.gap = cmd_buff[4];
-    u32 src_process_handle = cmd_buff[6];
-    LOG_DEBUG(Service_Y2R, "called image_size=0x%08X, transfer_unit=%hu, transfer_stride=%hu, "
-        "src_process_handle=0x%08X", conversion.src_V.image_size,
-        conversion.src_V.transfer_unit, conversion.src_V.gap, src_process_handle);
 
+    cmd_buff[0] = IPC::MakeHeader(0x12, 1, 0);
     cmd_buff[1] = RESULT_SUCCESS.raw;
+
+    LOG_DEBUG(Service_Y2R, "called image_size=0x%08X, transfer_unit=%hu, transfer_stride=%hu, src_process_handle=0x%08X",
+            conversion.src_V.image_size, conversion.src_V.transfer_unit, conversion.src_V.gap, cmd_buff[6]);
 }
 
 static void SetSendingYUYV(Service::Interface* self) {
@@ -192,12 +335,76 @@ static void SetSendingYUYV(Service::Interface* self) {
     conversion.src_YUYV.image_size = cmd_buff[2];
     conversion.src_YUYV.transfer_unit = cmd_buff[3];
     conversion.src_YUYV.gap = cmd_buff[4];
-    u32 src_process_handle = cmd_buff[6];
-    LOG_DEBUG(Service_Y2R, "called image_size=0x%08X, transfer_unit=%hu, transfer_stride=%hu, "
-        "src_process_handle=0x%08X", conversion.src_YUYV.image_size,
-        conversion.src_YUYV.transfer_unit, conversion.src_YUYV.gap, src_process_handle);
 
+    cmd_buff[0] = IPC::MakeHeader(0x13, 1, 0);
+    cmd_buff[1] = RESULT_SUCCESS.raw;
+
+    LOG_DEBUG(Service_Y2R, "called image_size=0x%08X, transfer_unit=%hu, transfer_stride=%hu, src_process_handle=0x%08X",
+            conversion.src_YUYV.image_size, conversion.src_YUYV.transfer_unit, conversion.src_YUYV.gap, cmd_buff[6]);
+}
+
+/**
+ * Y2R::IsFinishedSendingYuv service function
+ * Output:
+ *       1 : Result of the function, 0 on success, otherwise error code
+ *       2 : u8, 0 = Not Finished, 1 = Finished
+ */
+static void IsFinishedSendingYuv(Service::Interface* self) {
+    u32* cmd_buff = Kernel::GetCommandBuffer();
+
+    cmd_buff[0] = IPC::MakeHeader(0x14, 2, 0);
+    cmd_buff[1] = RESULT_SUCCESS.raw;
+    cmd_buff[2] = 1;
+
+    LOG_WARNING(Service_Y2R, "(STUBBED) called");
+}
+
+/**
+ * Y2R::IsFinishedSendingY service function
+ * Output:
+ *       1 : Result of the function, 0 on success, otherwise error code
+ *       2 : u8, 0 = Not Finished, 1 = Finished
+ */
+static void IsFinishedSendingY(Service::Interface* self) {
+    u32* cmd_buff = Kernel::GetCommandBuffer();
+
+    cmd_buff[0] = IPC::MakeHeader(0x15, 2, 0);
     cmd_buff[1] = RESULT_SUCCESS.raw;
+    cmd_buff[2] = 1;
+
+    LOG_WARNING(Service_Y2R, "(STUBBED) called");
+}
+
+/**
+ * Y2R::IsFinishedSendingU service function
+ * Output:
+ *       1 : Result of the function, 0 on success, otherwise error code
+ *       2 : u8, 0 = Not Finished, 1 = Finished
+ */
+static void IsFinishedSendingU(Service::Interface* self) {
+    u32* cmd_buff = Kernel::GetCommandBuffer();
+
+    cmd_buff[0] = IPC::MakeHeader(0x16, 2, 0);
+    cmd_buff[1] = RESULT_SUCCESS.raw;
+    cmd_buff[2] = 1;
+
+    LOG_WARNING(Service_Y2R, "(STUBBED) called");
+}
+
+/**
+ * Y2R::IsFinishedSendingV service function
+ * Output:
+ *       1 : Result of the function, 0 on success, otherwise error code
+ *       2 : u8, 0 = Not Finished, 1 = Finished
+ */
+static void IsFinishedSendingV(Service::Interface* self) {
+    u32* cmd_buff = Kernel::GetCommandBuffer();
+
+    cmd_buff[0] = IPC::MakeHeader(0x17, 2, 0);
+    cmd_buff[1] = RESULT_SUCCESS.raw;
+    cmd_buff[2] = 1;
+
+    LOG_WARNING(Service_Y2R, "(STUBBED) called");
 }
 
 static void SetReceiving(Service::Interface* self) {
@@ -207,27 +414,66 @@ static void SetReceiving(Service::Interface* self) {
     conversion.dst.image_size = cmd_buff[2];
     conversion.dst.transfer_unit = cmd_buff[3];
     conversion.dst.gap = cmd_buff[4];
-    u32 dst_process_handle = cmd_buff[6];
-    LOG_DEBUG(Service_Y2R, "called image_size=0x%08X, transfer_unit=%hu, transfer_stride=%hu, "
-        "dst_process_handle=0x%08X", conversion.dst.image_size,
-        conversion.dst.transfer_unit, conversion.dst.gap,
-        dst_process_handle);
 
+    cmd_buff[0] = IPC::MakeHeader(0x18, 1, 0);
     cmd_buff[1] = RESULT_SUCCESS.raw;
+
+    LOG_DEBUG(Service_Y2R, "called image_size=0x%08X, transfer_unit=%hu, transfer_stride=%hu, dst_process_handle=0x%08X",
+            conversion.dst.image_size, conversion.dst.transfer_unit, conversion.dst.gap, cmd_buff[6]);
+}
+
+/**
+ * Y2R::IsFinishedReceiving service function
+ * Output:
+ *       1 : Result of the function, 0 on success, otherwise error code
+ *       2 : u8, 0 = Not Finished, 1 = Finished
+ */
+static void IsFinishedReceiving(Service::Interface* self) {
+    u32* cmd_buff = Kernel::GetCommandBuffer();
+
+    cmd_buff[0] = IPC::MakeHeader(0x19, 2, 0);
+    cmd_buff[1] = RESULT_SUCCESS.raw;
+    cmd_buff[2] = 1;
+
+    LOG_WARNING(Service_Y2R, "(STUBBED) called");
 }
 
 static void SetInputLineWidth(Service::Interface* self) {
     u32* cmd_buff = Kernel::GetCommandBuffer();
 
-    LOG_DEBUG(Service_Y2R, "called input_line_width=%u", cmd_buff[1]);
+    cmd_buff[0] = IPC::MakeHeader(0x1A, 1, 0);
     cmd_buff[1] = conversion.SetInputLineWidth(cmd_buff[1]).raw;
+
+    LOG_DEBUG(Service_Y2R, "called input_line_width=%u", cmd_buff[1]);
+}
+
+static void GetInputLineWidth(Service::Interface* self) {
+    u32* cmd_buff = Kernel::GetCommandBuffer();
+
+    cmd_buff[0] = IPC::MakeHeader(0x1B, 2, 0);
+    cmd_buff[1] = RESULT_SUCCESS.raw;
+    cmd_buff[2] = conversion.input_line_width;
+
+    LOG_DEBUG(Service_Y2R, "called input_line_width=%u", conversion.input_line_width);
 }
 
 static void SetInputLines(Service::Interface* self) {
     u32* cmd_buff = Kernel::GetCommandBuffer();
 
-    LOG_DEBUG(Service_Y2R, "called input_line_number=%u", cmd_buff[1]);
+    cmd_buff[0] = IPC::MakeHeader(0x1C, 1, 0);
     cmd_buff[1] = conversion.SetInputLines(cmd_buff[1]).raw;
+
+    LOG_DEBUG(Service_Y2R, "called input_lines=%u", cmd_buff[1]);
+}
+
+static void GetInputLines(Service::Interface* self) {
+    u32* cmd_buff = Kernel::GetCommandBuffer();
+
+    cmd_buff[0] = IPC::MakeHeader(0x1D, 2, 0);
+    cmd_buff[1] = RESULT_SUCCESS.raw;
+    cmd_buff[2] = static_cast<u32>(conversion.input_lines);
+
+    LOG_DEBUG(Service_Y2R, "called input_lines=%u", conversion.input_lines);
 }
 
 static void SetCoefficient(Service::Interface* self) {
@@ -235,45 +481,111 @@ static void SetCoefficient(Service::Interface* self) {
 
     const u16* coefficients = reinterpret_cast<const u16*>(&cmd_buff[1]);
     std::memcpy(conversion.coefficients.data(), coefficients, sizeof(CoefficientSet));
+
+    cmd_buff[0] = IPC::MakeHeader(0x1E, 1, 0);
+    cmd_buff[1] = RESULT_SUCCESS.raw;
+
     LOG_DEBUG(Service_Y2R, "called coefficients=[%hX, %hX, %hX, %hX, %hX, %hX, %hX, %hX]",
             coefficients[0], coefficients[1], coefficients[2], coefficients[3],
             coefficients[4], coefficients[5], coefficients[6], coefficients[7]);
+}
 
+static void GetCoefficient(Service::Interface* self) {
+    u32* cmd_buff = Kernel::GetCommandBuffer();
+
+    cmd_buff[0] = IPC::MakeHeader(0x1F, 5, 0);
     cmd_buff[1] = RESULT_SUCCESS.raw;
+    std::memcpy(&cmd_buff[2], conversion.coefficients.data(), sizeof(CoefficientSet));
+
+    LOG_DEBUG(Service_Y2R, "called");
 }
 
 static void SetStandardCoefficient(Service::Interface* self) {
     u32* cmd_buff = Kernel::GetCommandBuffer();
 
-    LOG_DEBUG(Service_Y2R, "called standard_coefficient=%u", cmd_buff[1]);
+    u32 index = cmd_buff[1];
+
+    cmd_buff[0] = IPC::MakeHeader(0x20, 1, 0);
+    cmd_buff[1] = conversion.SetStandardCoefficient((StandardCoefficient)index).raw;
+
+    LOG_DEBUG(Service_Y2R, "called standard_coefficient=%u", index);
+}
+
+static void GetStandardCoefficient(Service::Interface* self) {
+    u32* cmd_buff = Kernel::GetCommandBuffer();
+
+    u32 index = cmd_buff[1];
+
+    if (index < ARRAY_SIZE(standard_coefficients)) {
+        cmd_buff[0] = IPC::MakeHeader(0x21, 5, 0);
+        cmd_buff[1] = RESULT_SUCCESS.raw;
+        std::memcpy(&cmd_buff[2], &standard_coefficients[index], sizeof(CoefficientSet));
 
-    cmd_buff[1] = conversion.SetStandardCoefficient((StandardCoefficient)cmd_buff[1]).raw;
+        LOG_DEBUG(Service_Y2R, "called standard_coefficient=%u ", index);
+    } else {
+        cmd_buff[0] = IPC::MakeHeader(0x21, 1, 0);
+        cmd_buff[1] = -1; // TODO(bunnei): Identify the correct error code for this
+
+        LOG_ERROR(Service_Y2R, "called standard_coefficient=%u  The argument is invalid!", index);
+    }
 }
 
 static void SetAlpha(Service::Interface* self) {
     u32* cmd_buff = Kernel::GetCommandBuffer();
 
     conversion.alpha = cmd_buff[1];
+
+    cmd_buff[0] = IPC::MakeHeader(0x22, 1, 0);
+    cmd_buff[1] = RESULT_SUCCESS.raw;
+
     LOG_DEBUG(Service_Y2R, "called alpha=%hu", conversion.alpha);
+}
+
+static void GetAlpha(Service::Interface* self) {
+    u32* cmd_buff = Kernel::GetCommandBuffer();
 
+    cmd_buff[0] = IPC::MakeHeader(0x23, 2, 0);
     cmd_buff[1] = RESULT_SUCCESS.raw;
+    cmd_buff[2] = conversion.alpha;
+
+    LOG_DEBUG(Service_Y2R, "called alpha=%hu", conversion.alpha);
 }
 
-static void StartConversion(Service::Interface* self) {
+static void SetDitheringWeightParams(Service::Interface* self) {
     u32* cmd_buff = Kernel::GetCommandBuffer();
+    std::memcpy(&dithering_weight_params, &cmd_buff[1], sizeof(DitheringWeightParams));
 
-    HW::Y2R::PerformConversion(conversion);
+    cmd_buff[0] = IPC::MakeHeader(0x24, 1, 0);
+    cmd_buff[1] = RESULT_SUCCESS.raw;
 
-    // dst_image_size would seem to be perfect for this, but it doesn't include the gap :(
-    u32 total_output_size = conversion.input_lines *
-        (conversion.dst.transfer_unit + conversion.dst.gap);
-    VideoCore::g_renderer->Rasterizer()->InvalidateRegion(
-        Memory::VirtualToPhysicalAddress(conversion.dst.address), total_output_size);
+    LOG_DEBUG(Service_Y2R, "called");
+}
+
+static void GetDitheringWeightParams(Service::Interface* self) {
+    u32* cmd_buff = Kernel::GetCommandBuffer();
+
+    cmd_buff[0] = IPC::MakeHeader(0x25, 9, 0);
+    cmd_buff[1] = RESULT_SUCCESS.raw;
+    std::memcpy(&cmd_buff[2], &dithering_weight_params, sizeof(DitheringWeightParams));
 
     LOG_DEBUG(Service_Y2R, "called");
+}
+
+static void StartConversion(Service::Interface* self) {
+    u32* cmd_buff = Kernel::GetCommandBuffer();
+
+    // dst_image_size would seem to be perfect for this, but it doesn't include the gap :(
+    u32 total_output_size = conversion.input_lines * (conversion.dst.transfer_unit + conversion.dst.gap);
+    Memory::RasterizerFlushAndInvalidateRegion(Memory::VirtualToPhysicalAddress(conversion.dst.address), total_output_size);
+
+    HW::Y2R::PerformConversion(conversion);
+
     completion_event->Signal();
 
+    cmd_buff[0] = IPC::MakeHeader(0x26, 1, 0);
     cmd_buff[1] = RESULT_SUCCESS.raw;
+
+    LOG_DEBUG(Service_Y2R, "called");
 }
 
 static void StopConversion(Service::Interface* self) {
@@ -281,6 +593,7 @@ static void StopConversion(Service::Interface* self) {
 
     cmd_buff[0] = IPC::MakeHeader(0x27, 1, 0);
     cmd_buff[1] = RESULT_SUCCESS.raw;
+
     LOG_DEBUG(Service_Y2R, "called");
 }
 
@@ -293,50 +606,61 @@ static void StopConversion(Service::Interface* self) {
 static void IsBusyConversion(Service::Interface* self) {
     u32* cmd_buff = Kernel::GetCommandBuffer();
 
+    cmd_buff[0] = IPC::MakeHeader(0x28, 2, 0);
     cmd_buff[1] = RESULT_SUCCESS.raw;
     cmd_buff[2] = 0; // StartConversion always finishes immediately
+
     LOG_DEBUG(Service_Y2R, "called");
 }
 
 /**
- * Y2R_U::SetConversionParams service function
+ * Y2R_U::SetPackageParameter service function
  */
-static void SetConversionParams(Service::Interface* self) {
+static void SetPackageParameter(Service::Interface* self) {
     u32* cmd_buff = Kernel::GetCommandBuffer();
 
     auto params = reinterpret_cast<const ConversionParameters*>(&cmd_buff[1]);
-    LOG_DEBUG(Service_Y2R,
-        "called input_format=%hhu output_format=%hhu rotation=%hhu block_alignment=%hhu "
-        "input_line_width=%hu input_lines=%hu standard_coefficient=%hhu "
-        "reserved=%hhu alpha=%hX",
-        params->input_format, params->output_format, params->rotation, params->block_alignment,
-        params->input_line_width, params->input_lines, params->standard_coefficient,
-        params->reserved, params->alpha);
-
-    ResultCode result = RESULT_SUCCESS;
 
     conversion.input_format = params->input_format;
     conversion.output_format = params->output_format;
     conversion.rotation = params->rotation;
     conversion.block_alignment = params->block_alignment;
-    result = conversion.SetInputLineWidth(params->input_line_width);
-    if (result.IsError()) goto cleanup;
+
+    ResultCode result = conversion.SetInputLineWidth(params->input_line_width);
+
+    if (result.IsError())
+        goto cleanup;
+
     result = conversion.SetInputLines(params->input_lines);
-    if (result.IsError()) goto cleanup;
+
+    if (result.IsError())
+        goto cleanup;
+
     result = conversion.SetStandardCoefficient(params->standard_coefficient);
-    if (result.IsError()) goto cleanup;
+
+    if (result.IsError())
+        goto cleanup;
+
+    conversion.padding = params->padding;
     conversion.alpha = params->alpha;
 
 cleanup:
     cmd_buff[0] = IPC::MakeHeader(0x29, 1, 0);
     cmd_buff[1] = result.raw;
+
+    LOG_DEBUG(Service_Y2R, "called input_format=%hhu output_format=%hhu rotation=%hhu block_alignment=%hhu "
+            "input_line_width=%hu input_lines=%hu standard_coefficient=%hhu reserved=%hhu alpha=%hX",
+            params->input_format, params->output_format, params->rotation, params->block_alignment,
+            params->input_line_width, params->input_lines, params->standard_coefficient, params->padding, params->alpha);
 }
 
 static void PingProcess(Service::Interface* self) {
     u32* cmd_buff = Kernel::GetCommandBuffer();
 
+    cmd_buff[0] = IPC::MakeHeader(0x2A, 2, 0);
     cmd_buff[1] = RESULT_SUCCESS.raw;
     cmd_buff[2] = 0;
+
     LOG_WARNING(Service_Y2R, "(STUBBED) called");
 }
 
@@ -362,6 +686,7 @@ static void DriverInitialize(Service::Interface* self) {
 
     cmd_buff[0] = IPC::MakeHeader(0x2B, 1, 0);
     cmd_buff[1] = RESULT_SUCCESS.raw;
+
     LOG_DEBUG(Service_Y2R, "called");
 }
 
@@ -370,54 +695,67 @@ static void DriverFinalize(Service::Interface* self) {
 
     cmd_buff[0] = IPC::MakeHeader(0x2C, 1, 0);
     cmd_buff[1] = RESULT_SUCCESS.raw;
+
+    LOG_DEBUG(Service_Y2R, "called");
+}
+
+
+static void GetPackageParameter(Service::Interface* self) {
+    u32* cmd_buff = Kernel::GetCommandBuffer();
+
+    cmd_buff[0] = IPC::MakeHeader(0x2D, 4, 0);
+    cmd_buff[1] = RESULT_SUCCESS.raw;
+    std::memcpy(&cmd_buff[2], &conversion, sizeof(ConversionParameters));
+
     LOG_DEBUG(Service_Y2R, "called");
 }
 
 const Interface::FunctionInfo FunctionTable[] = {
     {0x00010040, SetInputFormat,          "SetInputFormat"},
-    {0x00020000, nullptr,                 "GetInputFormat"},
+    {0x00020000, GetInputFormat,          "GetInputFormat"},
     {0x00030040, SetOutputFormat,         "SetOutputFormat"},
-    {0x00040000, nullptr,                 "GetOutputFormat"},
+    {0x00040000, GetOutputFormat,         "GetOutputFormat"},
     {0x00050040, SetRotation,             "SetRotation"},
-    {0x00060000, nullptr,                 "GetRotation"},
+    {0x00060000, GetRotation,             "GetRotation"},
     {0x00070040, SetBlockAlignment,       "SetBlockAlignment"},
-    {0x00080000, nullptr,                 "GetBlockAlignment"},
-    {0x00090040, nullptr,                 "SetSpacialDithering"},
-    {0x000A0000, nullptr,                 "GetSpacialDithering"},
-    {0x000B0040, nullptr,                 "SetTemporalDithering"},
-    {0x000C0000, nullptr,                 "GetTemporalDithering"},
+    {0x00080000, GetBlockAlignment,       "GetBlockAlignment"},
+    {0x00090040, SetSpacialDithering,     "SetSpacialDithering"},
+    {0x000A0000, GetSpacialDithering,     "GetSpacialDithering"},
+    {0x000B0040, SetTemporalDithering,    "SetTemporalDithering"},
+    {0x000C0000, GetTemporalDithering,    "GetTemporalDithering"},
     {0x000D0040, SetTransferEndInterrupt, "SetTransferEndInterrupt"},
+    {0x000E0000, GetTransferEndInterrupt, "GetTransferEndInterrupt"},
     {0x000F0000, GetTransferEndEvent,     "GetTransferEndEvent"},
     {0x00100102, SetSendingY,             "SetSendingY"},
     {0x00110102, SetSendingU,             "SetSendingU"},
     {0x00120102, SetSendingV,             "SetSendingV"},
     {0x00130102, SetSendingYUYV,          "SetSendingYUYV"},
-    {0x00140000, nullptr,                 "IsFinishedSendingYuv"},
-    {0x00150000, nullptr,                 "IsFinishedSendingY"},
-    {0x00160000, nullptr,                 "IsFinishedSendingU"},
-    {0x00170000, nullptr,                 "IsFinishedSendingV"},
+    {0x00140000, IsFinishedSendingYuv,    "IsFinishedSendingYuv"},
+    {0x00150000, IsFinishedSendingY,      "IsFinishedSendingY"},
+    {0x00160000, IsFinishedSendingU,      "IsFinishedSendingU"},
+    {0x00170000, IsFinishedSendingV,      "IsFinishedSendingV"},
     {0x00180102, SetReceiving,            "SetReceiving"},
-    {0x00190000, nullptr,                 "IsFinishedReceiving"},
+    {0x00190000, IsFinishedReceiving,     "IsFinishedReceiving"},
     {0x001A0040, SetInputLineWidth,       "SetInputLineWidth"},
-    {0x001B0000, nullptr,                 "GetInputLineWidth"},
+    {0x001B0000, GetInputLineWidth,       "GetInputLineWidth"},
     {0x001C0040, SetInputLines,           "SetInputLines"},
-    {0x001D0000, nullptr,                 "GetInputLines"},
+    {0x001D0000, GetInputLines,           "GetInputLines"},
     {0x001E0100, SetCoefficient,          "SetCoefficient"},
-    {0x001F0000, nullptr,                 "GetCoefficient"},
+    {0x001F0000, GetCoefficient,          "GetCoefficient"},
     {0x00200040, SetStandardCoefficient,  "SetStandardCoefficient"},
-    {0x00210040, nullptr,                 "GetStandardCoefficientParams"},
+    {0x00210040, GetStandardCoefficient,  "GetStandardCoefficient"},
     {0x00220040, SetAlpha,                "SetAlpha"},
-    {0x00230000, nullptr,                 "GetAlpha"},
-    {0x00240200, nullptr,                 "SetDitheringWeightParams"},
-    {0x00250000, nullptr,                 "GetDitheringWeightParams"},
+    {0x00230000, GetAlpha,                "GetAlpha"},
+    {0x00240200, SetDitheringWeightParams,"SetDitheringWeightParams"},
+    {0x00250000, GetDitheringWeightParams,"GetDitheringWeightParams"},
     {0x00260000, StartConversion,         "StartConversion"},
     {0x00270000, StopConversion,          "StopConversion"},
     {0x00280000, IsBusyConversion,        "IsBusyConversion"},
-    {0x002901C0, SetConversionParams,     "SetConversionParams"},
+    {0x002901C0, SetPackageParameter,     "SetPackageParameter"},
     {0x002A0000, PingProcess,             "PingProcess"},
     {0x002B0000, DriverInitialize,        "DriverInitialize"},
     {0x002C0000, DriverFinalize,          "DriverFinalize"},
-    {0x002D0000, nullptr,                 "GetPackageParameter"},
+    {0x002D0000, GetPackageParameter,     "GetPackageParameter"},
 };
 
 ////////////////////////////////////////////////////////////////////////////////////////////////////
diff --git a/src/core/hle/service/y2r_u.h b/src/core/hle/service/y2r_u.h
index 3965a5545..95fa2fdb7 100644
--- a/src/core/hle/service/y2r_u.h
+++ b/src/core/hle/service/y2r_u.h
@@ -97,6 +97,7 @@ struct ConversionConfiguration {
     u16 input_line_width;
     u16 input_lines;
     CoefficientSet coefficients;
+    u8 padding;
     u16 alpha;
 
     /// Input parameters for the Y (luma) plane
@@ -109,6 +110,25 @@ struct ConversionConfiguration {
     ResultCode SetStandardCoefficient(StandardCoefficient standard_coefficient);
 };
 
+struct DitheringWeightParams {
+    u16 w0_xEven_yEven;
+    u16 w0_xOdd_yEven;
+    u16 w0_xEven_yOdd;
+    u16 w0_xOdd_yOdd;
+    u16 w1_xEven_yEven;
+    u16 w1_xOdd_yEven;
+    u16 w1_xEven_yOdd;
+    u16 w1_xOdd_yOdd;
+    u16 w2_xEven_yEven;
+    u16 w2_xOdd_yEven;
+    u16 w2_xEven_yOdd;
+    u16 w2_xOdd_yOdd;
+    u16 w3_xEven_yEven;
+    u16 w3_xOdd_yEven;
+    u16 w3_xEven_yOdd;
+    u16 w3_xOdd_yOdd;
+};
+
 class Interface : public Service::Interface {
 public:
     Interface();
diff --git a/src/core/hle/svc.cpp b/src/core/hle/svc.cpp
index ae54afb1c..fb2aecbf2 100644
--- a/src/core/hle/svc.cpp
+++ b/src/core/hle/svc.cpp
@@ -6,7 +6,6 @@
 
 #include "common/logging/log.h"
 #include "common/microprofile.h"
-#include "common/profiler.h"
 #include "common/string_util.h"
 #include "common/symbols.h"
 
@@ -860,6 +859,10 @@ static ResultCode GetProcessInfo(s64* out, Handle process_handle, u32 type) {
         // TODO(yuriks): Type 0 returns a slightly higher number than type 2, but I'm not sure
         // what's the difference between them.
         *out = process->heap_used + process->linear_heap_used + process->misc_memory_used;
+        if(*out % Memory::PAGE_SIZE != 0) {
+            LOG_ERROR(Kernel_SVC, "called, memory size not page-aligned");
+            return ERR_MISALIGNED_SIZE;
+        }
         break;
     case 1:
     case 3:
@@ -1031,8 +1034,6 @@ static const FunctionDef SVC_Table[] = {
     {0x7D, HLE::Wrap<QueryProcessMemory>,   "QueryProcessMemory"},
 };
 
-Common::Profiling::TimingCategory profiler_svc("SVC Calls");
-
 static const FunctionDef* GetSVCInfo(u32 func_num) {
     if (func_num >= ARRAY_SIZE(SVC_Table)) {
         LOG_ERROR(Kernel_SVC, "unknown svc=0x%02X", func_num);
@@ -1044,7 +1045,6 @@ static const FunctionDef* GetSVCInfo(u32 func_num) {
 MICROPROFILE_DEFINE(Kernel_SVC, "Kernel", "SVC", MP_RGB(70, 200, 70));
 
 void CallSVC(u32 immediate) {
-    Common::Profiling::ScopeTimer timer_svc(profiler_svc);
     MICROPROFILE_SCOPE(Kernel_SVC);
 
     const FunctionDef* info = GetSVCInfo(immediate);
diff --git a/src/core/hw/gpu.cpp b/src/core/hw/gpu.cpp
index 7e2f9cdfa..2fe856293 100644
--- a/src/core/hw/gpu.cpp
+++ b/src/core/hw/gpu.cpp
@@ -115,21 +115,39 @@ inline void Write(u32 addr, const T data) {
                 u8* start = Memory::GetPhysicalPointer(config.GetStartAddress());
                 u8* end = Memory::GetPhysicalPointer(config.GetEndAddress());
 
-                if (config.fill_24bit) {
-                    // fill with 24-bit values
-                    for (u8* ptr = start; ptr < end; ptr += 3) {
-                        ptr[0] = config.value_24bit_r;
-                        ptr[1] = config.value_24bit_g;
-                        ptr[2] = config.value_24bit_b;
+                // TODO: Consider always accelerating and returning vector of
+                //       regions that the accelerated fill did not cover to
+                //       reduce/eliminate the fill that the cpu has to do.
+                //       This would also mean that the flush below is not needed.
+                //       Fill should first flush all surfaces that touch but are
+                //       not completely within the fill range.
+                //       Then fill all completely covered surfaces, and return the
+                //       regions that were between surfaces or within the touching
+                //       ones for cpu to manually fill here.
+                if (!VideoCore::g_renderer->Rasterizer()->AccelerateFill(config)) {
+                    Memory::RasterizerFlushAndInvalidateRegion(config.GetStartAddress(), config.GetEndAddress() - config.GetStartAddress());
+
+                    if (config.fill_24bit) {
+                        // fill with 24-bit values
+                        for (u8* ptr = start; ptr < end; ptr += 3) {
+                            ptr[0] = config.value_24bit_r;
+                            ptr[1] = config.value_24bit_g;
+                            ptr[2] = config.value_24bit_b;
+                        }
+                    } else if (config.fill_32bit) {
+                        // fill with 32-bit values
+                        if (end > start) {
+                            u32 value = config.value_32bit;
+                            size_t len = (end - start) / sizeof(u32);
+                            for (size_t i = 0; i < len; ++i)
+                                memcpy(&start[i * sizeof(u32)], &value, sizeof(u32));
+                        }
+                    } else {
+                        // fill with 16-bit values
+                        u16 value_16bit = config.value_16bit.Value();
+                        for (u8* ptr = start; ptr < end; ptr += sizeof(u16))
+                            memcpy(ptr, &value_16bit, sizeof(u16));
                     }
-                } else if (config.fill_32bit) {
-                    // fill with 32-bit values
-                    for (u32* ptr = (u32*)start; ptr < (u32*)end; ++ptr)
-                        *ptr = config.value_32bit;
-                } else {
-                    // fill with 16-bit values
-                    for (u16* ptr = (u16*)start; ptr < (u16*)end; ++ptr)
-                        *ptr = config.value_16bit;
                 }
 
                 LOG_TRACE(HW_GPU, "MemoryFill from 0x%08x to 0x%08x", config.GetStartAddress(), config.GetEndAddress());
@@ -139,8 +157,6 @@ inline void Write(u32 addr, const T data) {
                 } else {
                     GSP_GPU::SignalInterrupt(GSP_GPU::InterruptId::PSC1);
                 }
-
-                VideoCore::g_renderer->Rasterizer()->InvalidateRegion(config.GetStartAddress(), config.GetEndAddress() - config.GetStartAddress());
             }
 
             // Reset "trigger" flag and set the "finish" flag
@@ -161,184 +177,185 @@ inline void Write(u32 addr, const T data) {
             if (Pica::g_debug_context)
                 Pica::g_debug_context->OnEvent(Pica::DebugContext::Event::IncomingDisplayTransfer, nullptr);
 
-            u8* src_pointer = Memory::GetPhysicalPointer(config.GetPhysicalInputAddress());
-            u8* dst_pointer = Memory::GetPhysicalPointer(config.GetPhysicalOutputAddress());
-
-            if (config.is_texture_copy) {
-                u32 input_width = config.texture_copy.input_width * 16;
-                u32 input_gap = config.texture_copy.input_gap * 16;
-                u32 output_width = config.texture_copy.output_width * 16;
-                u32 output_gap = config.texture_copy.output_gap * 16;
-
-                size_t contiguous_input_size = config.texture_copy.size / input_width * (input_width + input_gap);
-                VideoCore::g_renderer->Rasterizer()->FlushRegion(config.GetPhysicalInputAddress(), contiguous_input_size);
-
-                u32 remaining_size = config.texture_copy.size;
-                u32 remaining_input = input_width;
-                u32 remaining_output = output_width;
-                while (remaining_size > 0) {
-                    u32 copy_size = std::min({ remaining_input, remaining_output, remaining_size });
+            if (!VideoCore::g_renderer->Rasterizer()->AccelerateDisplayTransfer(config)) {
+                u8* src_pointer = Memory::GetPhysicalPointer(config.GetPhysicalInputAddress());
+                u8* dst_pointer = Memory::GetPhysicalPointer(config.GetPhysicalOutputAddress());
 
-                    std::memcpy(dst_pointer, src_pointer, copy_size);
-                    src_pointer += copy_size;
-                    dst_pointer += copy_size;
+                if (config.is_texture_copy) {
+                    u32 input_width = config.texture_copy.input_width * 16;
+                    u32 input_gap = config.texture_copy.input_gap * 16;
+                    u32 output_width = config.texture_copy.output_width * 16;
+                    u32 output_gap = config.texture_copy.output_gap * 16;
 
-                    remaining_input -= copy_size;
-                    remaining_output -= copy_size;
-                    remaining_size -= copy_size;
+                    size_t contiguous_input_size = config.texture_copy.size / input_width * (input_width + input_gap);
+                    Memory::RasterizerFlushRegion(config.GetPhysicalInputAddress(), contiguous_input_size);
 
-                    if (remaining_input == 0) {
-                        remaining_input = input_width;
-                        src_pointer += input_gap;
-                    }
-                    if (remaining_output == 0) {
-                        remaining_output = output_width;
-                        dst_pointer += output_gap;
-                    }
-                }
+                    size_t contiguous_output_size = config.texture_copy.size / output_width * (output_width + output_gap);
+                    Memory::RasterizerFlushAndInvalidateRegion(config.GetPhysicalOutputAddress(), contiguous_output_size);
 
-                LOG_TRACE(HW_GPU, "TextureCopy: 0x%X bytes from 0x%08X(%u+%u)-> 0x%08X(%u+%u), flags 0x%08X",
-                    config.texture_copy.size,
-                    config.GetPhysicalInputAddress(), input_width, input_gap,
-                    config.GetPhysicalOutputAddress(), output_width, output_gap,
-                    config.flags);
+                    u32 remaining_size = config.texture_copy.size;
+                    u32 remaining_input = input_width;
+                    u32 remaining_output = output_width;
+                    while (remaining_size > 0) {
+                        u32 copy_size = std::min({ remaining_input, remaining_output, remaining_size });
 
-                size_t contiguous_output_size = config.texture_copy.size / output_width * (output_width + output_gap);
-                VideoCore::g_renderer->Rasterizer()->InvalidateRegion(config.GetPhysicalOutputAddress(), contiguous_output_size);
+                        std::memcpy(dst_pointer, src_pointer, copy_size);
+                        src_pointer += copy_size;
+                        dst_pointer += copy_size;
 
-                GSP_GPU::SignalInterrupt(GSP_GPU::InterruptId::PPF);
-                break;
-            }
+                        remaining_input -= copy_size;
+                        remaining_output -= copy_size;
+                        remaining_size -= copy_size;
 
-            if (config.scaling > config.ScaleXY) {
-                LOG_CRITICAL(HW_GPU, "Unimplemented display transfer scaling mode %u", config.scaling.Value());
-                UNIMPLEMENTED();
-                break;
-            }
+                        if (remaining_input == 0) {
+                            remaining_input = input_width;
+                            src_pointer += input_gap;
+                        }
+                        if (remaining_output == 0) {
+                            remaining_output = output_width;
+                            dst_pointer += output_gap;
+                        }
+                    }
 
-            if (config.input_linear && config.scaling != config.NoScale) {
-                LOG_CRITICAL(HW_GPU, "Scaling is only implemented on tiled input");
-                UNIMPLEMENTED();
-                break;
-            }
+                    LOG_TRACE(HW_GPU, "TextureCopy: 0x%X bytes from 0x%08X(%u+%u)-> 0x%08X(%u+%u), flags 0x%08X",
+                        config.texture_copy.size,
+                        config.GetPhysicalInputAddress(), input_width, input_gap,
+                        config.GetPhysicalOutputAddress(), output_width, output_gap,
+                        config.flags);
 
-            bool horizontal_scale = config.scaling != config.NoScale;
-            bool vertical_scale = config.scaling == config.ScaleXY;
+                    GSP_GPU::SignalInterrupt(GSP_GPU::InterruptId::PPF);
+                    break;
+                }
 
-            u32 output_width = config.output_width >> horizontal_scale;
-            u32 output_height = config.output_height >> vertical_scale;
+                if (config.scaling > config.ScaleXY) {
+                    LOG_CRITICAL(HW_GPU, "Unimplemented display transfer scaling mode %u", config.scaling.Value());
+                    UNIMPLEMENTED();
+                    break;
+                }
 
-            u32 input_size = config.input_width * config.input_height * GPU::Regs::BytesPerPixel(config.input_format);
-            u32 output_size = output_width * output_height * GPU::Regs::BytesPerPixel(config.output_format);
+                if (config.input_linear && config.scaling != config.NoScale) {
+                    LOG_CRITICAL(HW_GPU, "Scaling is only implemented on tiled input");
+                    UNIMPLEMENTED();
+                    break;
+                }
 
-            VideoCore::g_renderer->Rasterizer()->FlushRegion(config.GetPhysicalInputAddress(), input_size);
+                int horizontal_scale = config.scaling != config.NoScale ? 1 : 0;
+                int vertical_scale = config.scaling == config.ScaleXY ? 1 : 0;
 
-            for (u32 y = 0; y < output_height; ++y) {
-                for (u32 x = 0; x < output_width; ++x) {
-                    Math::Vec4<u8> src_color;
+                u32 output_width = config.output_width >> horizontal_scale;
+                u32 output_height = config.output_height >> vertical_scale;
 
-                    // Calculate the [x,y] position of the input image
-                    // based on the current output position and the scale
-                    u32 input_x = x << horizontal_scale;
-                    u32 input_y = y << vertical_scale;
+                u32 input_size = config.input_width * config.input_height * GPU::Regs::BytesPerPixel(config.input_format);
+                u32 output_size = output_width * output_height * GPU::Regs::BytesPerPixel(config.output_format);
 
-                    if (config.flip_vertically) {
-                        // Flip the y value of the output data,
-                        // we do this after calculating the [x,y] position of the input image
-                        // to account for the scaling options.
-                        y = output_height - y - 1;
-                    }
+                Memory::RasterizerFlushRegion(config.GetPhysicalInputAddress(), input_size);
+                Memory::RasterizerFlushAndInvalidateRegion(config.GetPhysicalOutputAddress(), output_size);
 
-                    u32 dst_bytes_per_pixel = GPU::Regs::BytesPerPixel(config.output_format);
-                    u32 src_bytes_per_pixel = GPU::Regs::BytesPerPixel(config.input_format);
-                    u32 src_offset;
-                    u32 dst_offset;
+                for (u32 y = 0; y < output_height; ++y) {
+                    for (u32 x = 0; x < output_width; ++x) {
+                        Math::Vec4<u8> src_color;
 
-                    if (config.input_linear) {
-                        if (!config.dont_swizzle) {
-                            // Interpret the input as linear and the output as tiled
-                            u32 coarse_y = y & ~7;
-                            u32 stride = output_width * dst_bytes_per_pixel;
+                        // Calculate the [x,y] position of the input image
+                        // based on the current output position and the scale
+                        u32 input_x = x << horizontal_scale;
+                        u32 input_y = y << vertical_scale;
 
-                            src_offset = (input_x + input_y * config.input_width) * src_bytes_per_pixel;
-                            dst_offset = VideoCore::GetMortonOffset(x, y, dst_bytes_per_pixel) + coarse_y * stride;
-                        } else {
-                           // Both input and output are linear
-                            src_offset = (input_x + input_y * config.input_width) * src_bytes_per_pixel;
-                            dst_offset = (x + y * output_width) * dst_bytes_per_pixel;
+                        if (config.flip_vertically) {
+                            // Flip the y value of the output data,
+                            // we do this after calculating the [x,y] position of the input image
+                            // to account for the scaling options.
+                            y = output_height - y - 1;
                         }
-                    } else {
-                        if (!config.dont_swizzle) {
-                            // Interpret the input as tiled and the output as linear
-                            u32 coarse_y = input_y & ~7;
-                            u32 stride = config.input_width * src_bytes_per_pixel;
 
-                            src_offset = VideoCore::GetMortonOffset(input_x, input_y, src_bytes_per_pixel) + coarse_y * stride;
-                            dst_offset = (x + y * output_width) * dst_bytes_per_pixel;
+                        u32 dst_bytes_per_pixel = GPU::Regs::BytesPerPixel(config.output_format);
+                        u32 src_bytes_per_pixel = GPU::Regs::BytesPerPixel(config.input_format);
+                        u32 src_offset;
+                        u32 dst_offset;
+
+                        if (config.input_linear) {
+                            if (!config.dont_swizzle) {
+                                // Interpret the input as linear and the output as tiled
+                                u32 coarse_y = y & ~7;
+                                u32 stride = output_width * dst_bytes_per_pixel;
+
+                                src_offset = (input_x + input_y * config.input_width) * src_bytes_per_pixel;
+                                dst_offset = VideoCore::GetMortonOffset(x, y, dst_bytes_per_pixel) + coarse_y * stride;
+                            } else {
+                                // Both input and output are linear
+                                src_offset = (input_x + input_y * config.input_width) * src_bytes_per_pixel;
+                                dst_offset = (x + y * output_width) * dst_bytes_per_pixel;
+                            }
                         } else {
-                            // Both input and output are tiled
-                            u32 out_coarse_y = y & ~7;
-                            u32 out_stride = output_width * dst_bytes_per_pixel;
-
-                            u32 in_coarse_y = input_y & ~7;
-                            u32 in_stride = config.input_width * src_bytes_per_pixel;
-
-                            src_offset = VideoCore::GetMortonOffset(input_x, input_y, src_bytes_per_pixel) + in_coarse_y * in_stride;
-                            dst_offset = VideoCore::GetMortonOffset(x, y, dst_bytes_per_pixel) + out_coarse_y * out_stride;
+                            if (!config.dont_swizzle) {
+                                // Interpret the input as tiled and the output as linear
+                                u32 coarse_y = input_y & ~7;
+                                u32 stride = config.input_width * src_bytes_per_pixel;
+
+                                src_offset = VideoCore::GetMortonOffset(input_x, input_y, src_bytes_per_pixel) + coarse_y * stride;
+                                dst_offset = (x + y * output_width) * dst_bytes_per_pixel;
+                            } else {
+                                // Both input and output are tiled
+                                u32 out_coarse_y = y & ~7;
+                                u32 out_stride = output_width * dst_bytes_per_pixel;
+
+                                u32 in_coarse_y = input_y & ~7;
+                                u32 in_stride = config.input_width * src_bytes_per_pixel;
+
+                                src_offset = VideoCore::GetMortonOffset(input_x, input_y, src_bytes_per_pixel) + in_coarse_y * in_stride;
+                                dst_offset = VideoCore::GetMortonOffset(x, y, dst_bytes_per_pixel) + out_coarse_y * out_stride;
+                            }
                         }
-                    }
 
-                    const u8* src_pixel = src_pointer + src_offset;
-                    src_color = DecodePixel(config.input_format, src_pixel);
-                    if (config.scaling == config.ScaleX) {
-                        Math::Vec4<u8> pixel = DecodePixel(config.input_format, src_pixel + src_bytes_per_pixel);
-                        src_color = ((src_color + pixel) / 2).Cast<u8>();
-                    } else if (config.scaling == config.ScaleXY) {
-                        Math::Vec4<u8> pixel1 = DecodePixel(config.input_format, src_pixel + 1 * src_bytes_per_pixel);
-                        Math::Vec4<u8> pixel2 = DecodePixel(config.input_format, src_pixel + 2 * src_bytes_per_pixel);
-                        Math::Vec4<u8> pixel3 = DecodePixel(config.input_format, src_pixel + 3 * src_bytes_per_pixel);
-                        src_color = (((src_color + pixel1) + (pixel2 + pixel3)) / 4).Cast<u8>();
-                    }
+                        const u8* src_pixel = src_pointer + src_offset;
+                        src_color = DecodePixel(config.input_format, src_pixel);
+                        if (config.scaling == config.ScaleX) {
+                            Math::Vec4<u8> pixel = DecodePixel(config.input_format, src_pixel + src_bytes_per_pixel);
+                            src_color = ((src_color + pixel) / 2).Cast<u8>();
+                        } else if (config.scaling == config.ScaleXY) {
+                            Math::Vec4<u8> pixel1 = DecodePixel(config.input_format, src_pixel + 1 * src_bytes_per_pixel);
+                            Math::Vec4<u8> pixel2 = DecodePixel(config.input_format, src_pixel + 2 * src_bytes_per_pixel);
+                            Math::Vec4<u8> pixel3 = DecodePixel(config.input_format, src_pixel + 3 * src_bytes_per_pixel);
+                            src_color = (((src_color + pixel1) + (pixel2 + pixel3)) / 4).Cast<u8>();
+                        }
 
-                    u8* dst_pixel = dst_pointer + dst_offset;
-                    switch (config.output_format) {
-                    case Regs::PixelFormat::RGBA8:
-                        Color::EncodeRGBA8(src_color, dst_pixel);
-                        break;
+                        u8* dst_pixel = dst_pointer + dst_offset;
+                        switch (config.output_format) {
+                        case Regs::PixelFormat::RGBA8:
+                            Color::EncodeRGBA8(src_color, dst_pixel);
+                            break;
 
-                    case Regs::PixelFormat::RGB8:
-                        Color::EncodeRGB8(src_color, dst_pixel);
-                        break;
+                        case Regs::PixelFormat::RGB8:
+                            Color::EncodeRGB8(src_color, dst_pixel);
+                            break;
 
-                    case Regs::PixelFormat::RGB565:
-                        Color::EncodeRGB565(src_color, dst_pixel);
-                        break;
+                        case Regs::PixelFormat::RGB565:
+                            Color::EncodeRGB565(src_color, dst_pixel);
+                            break;
 
-                    case Regs::PixelFormat::RGB5A1:
-                        Color::EncodeRGB5A1(src_color, dst_pixel);
-                        break;
+                        case Regs::PixelFormat::RGB5A1:
+                            Color::EncodeRGB5A1(src_color, dst_pixel);
+                            break;
 
-                    case Regs::PixelFormat::RGBA4:
-                        Color::EncodeRGBA4(src_color, dst_pixel);
-                        break;
+                        case Regs::PixelFormat::RGBA4:
+                            Color::EncodeRGBA4(src_color, dst_pixel);
+                            break;
 
-                    default:
-                        LOG_ERROR(HW_GPU, "Unknown destination framebuffer format %x", config.output_format.Value());
-                        break;
+                        default:
+                            LOG_ERROR(HW_GPU, "Unknown destination framebuffer format %x", config.output_format.Value());
+                            break;
+                        }
                     }
                 }
-            }
 
-            LOG_TRACE(HW_GPU, "DisplayTriggerTransfer: 0x%08x bytes from 0x%08x(%ux%u)-> 0x%08x(%ux%u), dst format %x, flags 0x%08X",
+                LOG_TRACE(HW_GPU, "DisplayTriggerTransfer: 0x%08x bytes from 0x%08x(%ux%u)-> 0x%08x(%ux%u), dst format %x, flags 0x%08X",
                       config.output_height * output_width * GPU::Regs::BytesPerPixel(config.output_format),
                       config.GetPhysicalInputAddress(), config.input_width.Value(), config.input_height.Value(),
                       config.GetPhysicalOutputAddress(), output_width, output_height,
                       config.output_format.Value(), config.flags);
+            }
 
             g_regs.display_transfer_config.trigger = 0;
             GSP_GPU::SignalInterrupt(GSP_GPU::InterruptId::PPF);
-
-            VideoCore::g_renderer->Rasterizer()->InvalidateRegion(config.GetPhysicalOutputAddress(), output_size);
         }
         break;
     }
diff --git a/src/core/hw/gpu.h b/src/core/hw/gpu.h
index a00adbf53..da4c345b4 100644
--- a/src/core/hw/gpu.h
+++ b/src/core/hw/gpu.h
@@ -78,7 +78,7 @@ struct Regs {
 
     INSERT_PADDING_WORDS(0x4);
 
-    struct {
+    struct MemoryFillConfig {
         u32 address_start;
         u32 address_end;
 
@@ -165,7 +165,7 @@ struct Regs {
 
     INSERT_PADDING_WORDS(0x169);
 
-    struct {
+    struct DisplayTransferConfig {
         u32 input_address;
         u32 output_address;
 
diff --git a/src/core/loader/ncch.cpp b/src/core/loader/ncch.cpp
index a4b47ef8c..066e91a9e 100644
--- a/src/core/loader/ncch.cpp
+++ b/src/core/loader/ncch.cpp
@@ -255,7 +255,7 @@ ResultStatus AppLoader_NCCH::Load() {
     resource_limit_category = exheader_header.arm11_system_local_caps.resource_limit_category;
 
     LOG_INFO(Loader,  "Name:                        %s"    , exheader_header.codeset_info.name);
-    LOG_INFO(Loader,  "Program ID:                  %016X" , ncch_header.program_id);
+    LOG_INFO(Loader,  "Program ID:                  %016llX" , ncch_header.program_id);
     LOG_DEBUG(Loader, "Code compressed:             %s"    , is_compressed ? "yes" : "no");
     LOG_DEBUG(Loader, "Entry point:                 0x%08X", entry_point);
     LOG_DEBUG(Loader, "Code size:                   0x%08X", code_size);
diff --git a/src/core/memory.cpp b/src/core/memory.cpp
index 7de5bd15d..ee9b69f81 100644
--- a/src/core/memory.cpp
+++ b/src/core/memory.cpp
@@ -15,6 +15,9 @@
 #include "core/memory_setup.h"
 #include "core/mmio.h"
 
+#include "video_core/renderer_base.h"
+#include "video_core/video_core.h"
+
 namespace Memory {
 
 enum class PageType {
@@ -22,8 +25,12 @@ enum class PageType {
     Unmapped,
     /// Page is mapped to regular memory. This is the only type you can get pointers to.
     Memory,
+    /// Page is mapped to regular memory, but also needs to check for rasterizer cache flushing and invalidation
+    RasterizerCachedMemory,
     /// Page is mapped to a I/O region. Writing and reading to this page is handled by functions.
     Special,
+    /// Page is mapped to a I/O region, but also needs to check for rasterizer cache flushing and invalidation
+    RasterizerCachedSpecial,
 };
 
 struct SpecialRegion {
@@ -57,6 +64,12 @@ struct PageTable {
      * the corresponding entry in `pointers` MUST be set to null.
      */
     std::array<PageType, NUM_ENTRIES> attributes;
+
+    /**
+     * Indicates the number of externally cached resources touching a page that should be
+     * flushed before the memory is accessed
+     */
+    std::array<u8, NUM_ENTRIES> cached_res_count;
 };
 
 /// Singular page table used for the singleton process
@@ -72,8 +85,15 @@ static void MapPages(u32 base, u32 size, u8* memory, PageType type) {
     while (base != end) {
         ASSERT_MSG(base < PageTable::NUM_ENTRIES, "out of range mapping at %08X", base);
 
+        // Since pages are unmapped on shutdown after video core is shutdown, the renderer may be null here
+        if (current_page_table->attributes[base] == PageType::RasterizerCachedMemory ||
+            current_page_table->attributes[base] == PageType::RasterizerCachedSpecial) {
+            RasterizerFlushAndInvalidateRegion(VirtualToPhysicalAddress(base << PAGE_BITS), PAGE_SIZE);
+        }
+
         current_page_table->attributes[base] = type;
         current_page_table->pointers[base] = memory;
+        current_page_table->cached_res_count[base] = 0;
 
         base += 1;
         if (memory != nullptr)
@@ -84,6 +104,7 @@ static void MapPages(u32 base, u32 size, u8* memory, PageType type) {
 void InitMemoryMap() {
     main_page_table.pointers.fill(nullptr);
     main_page_table.attributes.fill(PageType::Unmapped);
+    main_page_table.cached_res_count.fill(0);
 }
 
 void MapMemoryRegion(VAddr base, u32 size, u8* target) {
@@ -107,6 +128,28 @@ void UnmapRegion(VAddr base, u32 size) {
 }
 
 /**
+ * Gets a pointer to the exact memory at the virtual address (i.e. not page aligned)
+ * using a VMA from the current process
+ */
+static u8* GetPointerFromVMA(VAddr vaddr) {
+    u8* direct_pointer = nullptr;
+
+    auto& vma = Kernel::g_current_process->vm_manager.FindVMA(vaddr)->second;
+    switch (vma.type) {
+    case Kernel::VMAType::AllocatedMemoryBlock:
+        direct_pointer = vma.backing_block->data() + vma.offset;
+        break;
+    case Kernel::VMAType::BackingMemory:
+        direct_pointer = vma.backing_memory;
+        break;
+    default:
+        UNREACHABLE();
+    }
+
+    return direct_pointer + (vaddr - vma.base);
+}
+
+/**
  * This function should only be called for virtual addreses with attribute `PageType::Special`.
  */
 static MMIORegionPointer GetMMIOHandler(VAddr vaddr) {
@@ -126,6 +169,7 @@ template <typename T>
 T Read(const VAddr vaddr) {
     const u8* page_pointer = current_page_table->pointers[vaddr >> PAGE_BITS];
     if (page_pointer) {
+        // NOTE: Avoid adding any extra logic to this fast-path block
         T value;
         std::memcpy(&value, &page_pointer[vaddr & PAGE_MASK], sizeof(T));
         return value;
@@ -139,8 +183,22 @@ T Read(const VAddr vaddr) {
     case PageType::Memory:
         ASSERT_MSG(false, "Mapped memory page without a pointer @ %08X", vaddr);
         break;
+    case PageType::RasterizerCachedMemory:
+    {
+        RasterizerFlushRegion(VirtualToPhysicalAddress(vaddr), sizeof(T));
+
+        T value;
+        std::memcpy(&value, GetPointerFromVMA(vaddr), sizeof(T));
+        return value;
+    }
     case PageType::Special:
         return ReadMMIO<T>(GetMMIOHandler(vaddr), vaddr);
+    case PageType::RasterizerCachedSpecial:
+    {
+        RasterizerFlushRegion(VirtualToPhysicalAddress(vaddr), sizeof(T));
+
+        return ReadMMIO<T>(GetMMIOHandler(vaddr), vaddr);
+    }
     default:
         UNREACHABLE();
     }
@@ -153,6 +211,7 @@ template <typename T>
 void Write(const VAddr vaddr, const T data) {
     u8* page_pointer = current_page_table->pointers[vaddr >> PAGE_BITS];
     if (page_pointer) {
+        // NOTE: Avoid adding any extra logic to this fast-path block
         std::memcpy(&page_pointer[vaddr & PAGE_MASK], &data, sizeof(T));
         return;
     }
@@ -165,9 +224,23 @@ void Write(const VAddr vaddr, const T data) {
     case PageType::Memory:
         ASSERT_MSG(false, "Mapped memory page without a pointer @ %08X", vaddr);
         break;
+    case PageType::RasterizerCachedMemory:
+    {
+        RasterizerFlushAndInvalidateRegion(VirtualToPhysicalAddress(vaddr), sizeof(T));
+
+        std::memcpy(GetPointerFromVMA(vaddr), &data, sizeof(T));
+        break;
+    }
     case PageType::Special:
         WriteMMIO<T>(GetMMIOHandler(vaddr), vaddr, data);
         break;
+    case PageType::RasterizerCachedSpecial:
+    {
+        RasterizerFlushAndInvalidateRegion(VirtualToPhysicalAddress(vaddr), sizeof(T));
+
+        WriteMMIO<T>(GetMMIOHandler(vaddr), vaddr, data);
+        break;
+    }
     default:
         UNREACHABLE();
     }
@@ -179,6 +252,10 @@ u8* GetPointer(const VAddr vaddr) {
         return page_pointer + (vaddr & PAGE_MASK);
     }
 
+    if (current_page_table->attributes[vaddr >> PAGE_BITS] == PageType::RasterizerCachedMemory) {
+        return GetPointerFromVMA(vaddr);
+    }
+
     LOG_ERROR(HW_Memory, "unknown GetPointer @ 0x%08x", vaddr);
     return nullptr;
 }
@@ -187,6 +264,69 @@ u8* GetPhysicalPointer(PAddr address) {
     return GetPointer(PhysicalToVirtualAddress(address));
 }
 
+void RasterizerMarkRegionCached(PAddr start, u32 size, int count_delta) {
+    if (start == 0) {
+        return;
+    }
+
+    u32 num_pages = ((start + size - 1) >> PAGE_BITS) - (start >> PAGE_BITS) + 1;
+    PAddr paddr = start;
+
+    for (unsigned i = 0; i < num_pages; ++i) {
+        VAddr vaddr = PhysicalToVirtualAddress(paddr);
+        u8& res_count = current_page_table->cached_res_count[vaddr >> PAGE_BITS];
+        ASSERT_MSG(count_delta <= UINT8_MAX - res_count, "Rasterizer resource cache counter overflow!");
+        ASSERT_MSG(count_delta >= -res_count, "Rasterizer resource cache counter underflow!");
+
+        // Switch page type to cached if now cached
+        if (res_count == 0) {
+            PageType& page_type = current_page_table->attributes[vaddr >> PAGE_BITS];
+            switch (page_type) {
+            case PageType::Memory:
+                page_type = PageType::RasterizerCachedMemory;
+                current_page_table->pointers[vaddr >> PAGE_BITS] = nullptr;
+                break;
+            case PageType::Special:
+                page_type = PageType::RasterizerCachedSpecial;
+                break;
+            default:
+                UNREACHABLE();
+            }
+        }
+
+        res_count += count_delta;
+
+        // Switch page type to uncached if now uncached
+        if (res_count == 0) {
+            PageType& page_type = current_page_table->attributes[vaddr >> PAGE_BITS];
+            switch (page_type) {
+            case PageType::RasterizerCachedMemory:
+                page_type = PageType::Memory;
+                current_page_table->pointers[vaddr >> PAGE_BITS] = GetPointerFromVMA(vaddr & ~PAGE_MASK);
+                break;
+            case PageType::RasterizerCachedSpecial:
+                page_type = PageType::Special;
+                break;
+            default:
+                UNREACHABLE();
+            }
+        }
+        paddr += PAGE_SIZE;
+    }
+}
+
+void RasterizerFlushRegion(PAddr start, u32 size) {
+    if (VideoCore::g_renderer != nullptr) {
+        VideoCore::g_renderer->Rasterizer()->FlushRegion(start, size);
+    }
+}
+
+void RasterizerFlushAndInvalidateRegion(PAddr start, u32 size) {
+    if (VideoCore::g_renderer != nullptr) {
+        VideoCore::g_renderer->Rasterizer()->FlushAndInvalidateRegion(start, size);
+    }
+}
+
 u8 Read8(const VAddr addr) {
     return Read<u8>(addr);
 }
diff --git a/src/core/memory.h b/src/core/memory.h
index 5af72b7a7..9caa3c3f5 100644
--- a/src/core/memory.h
+++ b/src/core/memory.h
@@ -148,4 +148,20 @@ VAddr PhysicalToVirtualAddress(PAddr addr);
  */
 u8* GetPhysicalPointer(PAddr address);
 
+/**
+ * Adds the supplied value to the rasterizer resource cache counter of each
+ * page touching the region.
+ */
+void RasterizerMarkRegionCached(PAddr start, u32 size, int count_delta);
+
+/**
+ * Flushes any externally cached rasterizer resources touching the given region.
+ */
+void RasterizerFlushRegion(PAddr start, u32 size);
+
+/**
+ * Flushes and invalidates any externally cached rasterizer resources touching the given region.
+ */
+void RasterizerFlushAndInvalidateRegion(PAddr start, u32 size);
+
 }
diff --git a/src/core/settings.cpp b/src/core/settings.cpp
index 1aa26fbd2..eaf5c8461 100644
--- a/src/core/settings.cpp
+++ b/src/core/settings.cpp
@@ -19,7 +19,7 @@ void Apply() {
 
     VideoCore::g_hw_renderer_enabled = values.use_hw_renderer;
     VideoCore::g_shader_jit_enabled = values.use_shader_jit;
-
+    VideoCore::g_scaled_resolution_enabled = values.use_scaled_resolution;
 }
 
 } // namespace
diff --git a/src/core/settings.h b/src/core/settings.h
index 4933a516d..d620d8461 100644
--- a/src/core/settings.h
+++ b/src/core/settings.h
@@ -55,6 +55,7 @@ struct Values {
     // Renderer
     bool use_hw_renderer;
     bool use_shader_jit;
+    bool use_scaled_resolution;
 
     float bg_red;
     float bg_green;