110 files changed, 1811 insertions, 2244 deletions
diff --git a/src/common/CMakeLists.txt b/src/common/CMakeLists.txt
index 32cb85de0..f49a31612 100644
--- a/src/common/CMakeLists.txt
+++ b/src/common/CMakeLists.txt
@@ -31,7 +31,6 @@ add_library(common STATIC
    bit_set.h
    break_points.cpp
    break_points.h
-    chunk_file.h
    cityhash.cpp
    cityhash.h
    color.h
@@ -41,7 +40,6 @@ add_library(common STATIC
    file_util.cpp
    file_util.h
    hash.h
-    linear_disk_cache.h
    logging/backend.cpp
    logging/backend.h
    logging/filter.cpp
diff --git a/src/common/chunk_file.h b/src/common/chunk_file.h
deleted file mode 100644
index 972ef9039..000000000
--- a/src/common/chunk_file.h
+++ /dev/null
@@ -1,623 +0,0 @@
-// Copyright (C) 2003 Dolphin Project.
-// This program is free software: you can redistribute it and/or modify
-// it under the terms of the GNU General Public License as published by
-// the Free Software Foundation, version 2.0 or later versions.
-// This program is distributed in the hope that it will be useful,
-// but WITHOUT ANY WARRANTY; without even the implied warranty of
-// MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
-// GNU General Public License 2.0 for more details.
-// A copy of the GPL 2.0 should have been included with the program.
-// If not, see http://www.gnu.org/licenses/
-// Official SVN repository and contact information can be found at
-// http://code.google.com/p/dolphin-emu/
-#pragma once
-// Extremely simple serialization framework.
-// (mis)-features:
-// + Super fast
-// + Very simple
-// + Same code is used for serialization and deserializaition (in most cases)
-// - Zero backwards/forwards compatibility
-// - Serialization code for anything complex has to be manually written.
-#include <cstring>
-#include <deque>
-#include <list>
-#include <map>
-#include <set>
-#include <string>
-#include <type_traits>
-#include <utility>
-#include <vector>
-#include "common/assert.h"
-#include "common/common_types.h"
-#include "common/logging/log.h"
-template <class T>
-struct LinkedListItem : public T {
-    LinkedListItem<T>* next;
-};
-class PointerWrap;
-class PointerWrapSection {
-public:
-    PointerWrapSection(PointerWrap& p, int ver, const char* title)
-        : p_(p), ver_(ver), title_(title) {}
-    ~PointerWrapSection();
-    bool operator==(const int& v) const {
-        return ver_ == v;
-    }
-    bool operator!=(const int& v) const {
-        return ver_ != v;
-    }
-    bool operator<=(const int& v) const {
-        return ver_ <= v;
-    }
-    bool operator>=(const int& v) const {
-        return ver_ >= v;
-    }
-    bool operator<(const int& v) const {
-        return ver_ < v;
-    }
-    bool operator>(const int& v) const {
-        return ver_ > v;
-    }
-    operator bool() const {
-        return ver_ > 0;
-    }
-private:
-    PointerWrap& p_;
-    int ver_;
-    const char* title_;
-};
-// Wrapper class
-class PointerWrap {
-// This makes it a compile error if you forget to define DoState() on non-POD.
-// Which also can be a problem, for example struct tm is non-POD on linux, for whatever reason...
-#ifdef _MSC_VER
-    template <typename T, bool isPOD = std::is_pod<T>::value,
-              bool isPointer = std::is_pointer<T>::value>
-#else
-    template <typename T, bool isPOD = __is_pod(T), bool isPointer = std::is_pointer<T>::value>
-#endif
-    struct DoHelper {
-        static void DoArray(PointerWrap* p, T* x, int count) {
-            for (int i = 0; i < count; ++i)
-                p->Do(x[i]);
-        }
-        static void Do(PointerWrap* p, T& x) {
-            p->DoClass(x);
-        }
-    };
-    template <typename T>
-    struct DoHelper<T, true, false> {
-        static void DoArray(PointerWrap* p, T* x, int count) {
-            p->DoVoid((void*)x, sizeof(T) * count);
-        }
-        static void Do(PointerWrap* p, T& x) {
-            p->DoVoid((void*)&x, sizeof(x));
-        }
-    };
-public:
-    enum Mode {
-        MODE_READ = 1, // load
-        MODE_WRITE,    // save
-        MODE_MEASURE,  // calculate size
-        MODE_VERIFY,   // compare
-    };
-    enum Error {
-        ERROR_NONE = 0,
-        ERROR_WARNING = 1,
-        ERROR_FAILURE = 2,
-    };
-    u8** ptr;
-    Mode mode;
-    Error error;
-public:
-    PointerWrap(u8** ptr_, Mode mode_) : ptr(ptr_), mode(mode_), error(ERROR_NONE) {}
-    PointerWrap(unsigned char** ptr_, int mode_)
-        : ptr((u8**)ptr_), mode((Mode)mode_), error(ERROR_NONE) {}
-    PointerWrapSection Section(const char* title, int ver) {
-        return Section(title, ver, ver);
-    }
-    // The returned object can be compared against the version that was loaded.
-    // This can be used to support versions as old as minVer.
-    // Version = 0 means the section was not found.
-    PointerWrapSection Section(const char* title, int minVer, int ver) {
-        char marker[16] = {0};
-        int foundVersion = ver;
-        strncpy(marker, title, sizeof(marker));
-        if (!ExpectVoid(marker, sizeof(marker))) {
-            // Might be before we added name markers for safety.
-            if (foundVersion == 1 && ExpectVoid(&foundVersion, sizeof(foundVersion)))
-                DoMarker(title);
-            // Wasn't found, but maybe we can still load the state.
-            else
-                foundVersion = 0;
-        } else
-            Do(foundVersion);
-        if (error == ERROR_FAILURE || foundVersion < minVer || foundVersion > ver) {
-            LOG_ERROR(Common, "Savestate failure: wrong version %d found for %s", foundVersion,
-                      title);
-            SetError(ERROR_FAILURE);
-            return PointerWrapSection(*this, -1, title);
-        }
-        return PointerWrapSection(*this, foundVersion, title);
-    }
-    void SetMode(Mode mode_) {
-        mode = mode_;
-    }
-    Mode GetMode() const {
-        return mode;
-    }
-    u8** GetPPtr() {
-        return ptr;
-    }
-    void SetError(Error error_) {
-        if (error < error_)
-            error = error_;
-        if (error > ERROR_WARNING)
-            mode = PointerWrap::MODE_MEASURE;
-    }
-    bool ExpectVoid(void* data, int size) {
-        switch (mode) {
-        case MODE_READ:
-            if (memcmp(data, *ptr, size) != 0)
-                return false;
-            break;
-        case MODE_WRITE:
-            memcpy(*ptr, data, size);
-            break;
-        case MODE_MEASURE:
-            break; // MODE_MEASURE - don't need to do anything
-        case MODE_VERIFY:
-            for (int i = 0; i < size; i++) {
-                DEBUG_ASSERT_MSG(
-                    ((u8*)data)[i] == (*ptr)[i],
-                    "Savestate verification failure: %d (0x%X) (at %p) != %d (0x%X) (at %p).\n",
-                    ((u8*)data)[i], ((u8*)data)[i], &((u8*)data)[i], (*ptr)[i], (*ptr)[i],
-                    &(*ptr)[i]);
-            }
-            break;
-        default:
-            break; // throw an error?
-        }
-        (*ptr) += size;
-        return true;
-    }
-    void DoVoid(void* data, int size) {
-        switch (mode) {
-        case MODE_READ:
-            memcpy(data, *ptr, size);
-            break;
-        case MODE_WRITE:
-            memcpy(*ptr, data, size);
-            break;
-        case MODE_MEASURE:
-            break; // MODE_MEASURE - don't need to do anything
-        case MODE_VERIFY:
-            for (int i = 0; i < size; i++) {
-                DEBUG_ASSERT_MSG(
-                    ((u8*)data)[i] == (*ptr)[i],
-                    "Savestate verification failure: %d (0x%X) (at %p) != %d (0x%X) (at %p).\n",
-                    ((u8*)data)[i], ((u8*)data)[i], &((u8*)data)[i], (*ptr)[i], (*ptr)[i],
-                    &(*ptr)[i]);
-            }
-            break;
-        default:
-            break; // throw an error?
-        }
-        (*ptr) += size;
-    }
-    template <class K, class T>
-    void Do(std::map<K, T*>& x) {
-        if (mode == MODE_READ) {
-            for (auto it = x.begin(), end = x.end(); it != end; ++it) {
-                if (it->second != nullptr)
-                    delete it->second;
-            }
-        }
-        T* dv = nullptr;
-        DoMap(x, dv);
-    }
-    template <class K, class T>
-    void Do(std::map<K, T>& x) {
-        T dv = T();
-        DoMap(x, dv);
-    }
-    template <class K, class T>
-    void DoMap(std::map<K, T>& x, T& default_val) {
-        unsigned int number = (unsigned int)x.size();
-        Do(number);
-        switch (mode) {
-        case MODE_READ: {
-            x.clear();
-            while (number > 0) {
-                K first = K();
-                Do(first);
-                T second = default_val;
-                Do(second);
-                x[first] = second;
-                --number;
-            }
-        } break;
-        case MODE_WRITE:
-        case MODE_MEASURE:
-        case MODE_VERIFY: {
-            typename std::map<K, T>::iterator itr = x.begin();
-            while (number > 0) {
-                K first = itr->first;
-                Do(first);
-                Do(itr->second);
-                --number;
-                ++itr;
-            }
-        } break;
-        }
-    }
-    template <class K, class T>
-    void Do(std::multimap<K, T*>& x) {
-        if (mode == MODE_READ) {
-            for (auto it = x.begin(), end = x.end(); it != end; ++it) {
-                if (it->second != nullptr)
-                    delete it->second;
-            }
-        }
-        T* dv = nullptr;
-        DoMultimap(x, dv);
-    }
-    template <class K, class T>
-    void Do(std::multimap<K, T>& x) {
-        T dv = T();
-        DoMultimap(x, dv);
-    }
-    template <class K, class T>
-    void DoMultimap(std::multimap<K, T>& x, T& default_val) {
-        unsigned int number = (unsigned int)x.size();
-        Do(number);
-        switch (mode) {
-        case MODE_READ: {
-            x.clear();
-            while (number > 0) {
-                K first = K();
-                Do(first);
-                T second = default_val;
-                Do(second);
-                x.insert(std::make_pair(first, second));
-                --number;
-            }
-        } break;
-        case MODE_WRITE:
-        case MODE_MEASURE:
-        case MODE_VERIFY: {
-            typename std::multimap<K, T>::iterator itr = x.begin();
-            while (number > 0) {
-                Do(itr->first);
-                Do(itr->second);
-                --number;
-                ++itr;
-            }
-        } break;
-        }
-    }
-    // Store vectors.
-    template <class T>
-    void Do(std::vector<T*>& x) {
-        T* dv = nullptr;
-        DoVector(x, dv);
-    }
-    template <class T>
-    void Do(std::vector<T>& x) {
-        T dv = T();
-        DoVector(x, dv);
-    }
-    template <class T>
-    void DoPOD(std::vector<T>& x) {
-        T dv = T();
-        DoVectorPOD(x, dv);
-    }
-    template <class T>
-    void Do(std::vector<T>& x, T& default_val) {
-        DoVector(x, default_val);
-    }
-    template <class T>
-    void DoVector(std::vector<T>& x, T& default_val) {
-        u32 vec_size = (u32)x.size();
-        Do(vec_size);
-        x.resize(vec_size, default_val);
-        if (vec_size > 0)
-            DoArray(&x[0], vec_size);
-    }
-    template <class T>
-    void DoVectorPOD(std::vector<T>& x, T& default_val) {
-        u32 vec_size = (u32)x.size();
-        Do(vec_size);
-        x.resize(vec_size, default_val);
-        if (vec_size > 0)
-            DoArray(&x[0], vec_size);
-    }
-    // Store deques.
-    template <class T>
-    void Do(std::deque<T*>& x) {
-        T* dv = nullptr;
-        DoDeque(x, dv);
-    }
-    template <class T>
-    void Do(std::deque<T>& x) {
-        T dv = T();
-        DoDeque(x, dv);
-    }
-    template <class T>
-    void DoDeque(std::deque<T>& x, T& default_val) {
-        u32 deq_size = (u32)x.size();
-        Do(deq_size);
-        x.resize(deq_size, default_val);
-        u32 i;
-        for (i = 0; i < deq_size; i++)
-            Do(x[i]);
-    }
-    // Store STL lists.
-    template <class T>
-    void Do(std::list<T*>& x) {
-        T* dv = nullptr;
-        Do(x, dv);
-    }
-    template <class T>
-    void Do(std::list<T>& x) {
-        T dv = T();
-        DoList(x, dv);
-    }
-    template <class T>
-    void Do(std::list<T>& x, T& default_val) {
-        DoList(x, default_val);
-    }
-    template <class T>
-    void DoList(std::list<T>& x, T& default_val) {
-        u32 list_size = (u32)x.size();
-        Do(list_size);
-        x.resize(list_size, default_val);
-        typename std::list<T>::iterator itr, end;
-        for (itr = x.begin(), end = x.end(); itr != end; ++itr)
-            Do(*itr);
-    }
-    // Store STL sets.
-    template <class T>
-    void Do(std::set<T*>& x) {
-        if (mode == MODE_READ) {
-            for (auto it = x.begin(), end = x.end(); it != end; ++it) {
-                if (*it != nullptr)
-                    delete *it;
-            }
-        }
-        DoSet(x);
-    }
-    template <class T>
-    void Do(std::set<T>& x) {
-        DoSet(x);
-    }
-    template <class T>
-    void DoSet(std::set<T>& x) {
-        unsigned int number = (unsigned int)x.size();
-        Do(number);
-        switch (mode) {
-        case MODE_READ: {
-            x.clear();
-            while (number-- > 0) {
-                T it = T();
-                Do(it);
-                x.insert(it);
-            }
-        } break;
-        case MODE_WRITE:
-        case MODE_MEASURE:
-        case MODE_VERIFY: {
-            typename std::set<T>::iterator itr = x.begin();
-            while (number-- > 0)
-                Do(*itr++);
-        } break;
-        default:
-            LOG_ERROR(Common, "Savestate error: invalid mode %d.", mode);
-        }
-    }
-    // Store strings.
-    void Do(std::string& x) {
-        int stringLen = (int)x.length() + 1;
-        Do(stringLen);
-        switch (mode) {
-        case MODE_READ:
-            x = (char*)*ptr;
-            break;
-        case MODE_WRITE:
-            memcpy(*ptr, x.c_str(), stringLen);
-            break;
-        case MODE_MEASURE:
-            break;
-        case MODE_VERIFY:
-            DEBUG_ASSERT_MSG((x == (char*)*ptr),
-                             "Savestate verification failure: \"%s\" != \"%s\" (at %p).\n",
-                             x.c_str(), (char*)*ptr, ptr);
-            break;
-        }
-        (*ptr) += stringLen;
-    }
-    void Do(std::wstring& x) {
-        int stringLen = sizeof(wchar_t) * ((int)x.length() + 1);
-        Do(stringLen);
-        switch (mode) {
-        case MODE_READ:
-            x = (wchar_t*)*ptr;
-            break;
-        case MODE_WRITE:
-            memcpy(*ptr, x.c_str(), stringLen);
-            break;
-        case MODE_MEASURE:
-            break;
-        case MODE_VERIFY:
-            DEBUG_ASSERT_MSG((x == (wchar_t*)*ptr),
-                             "Savestate verification failure: \"%ls\" != \"%ls\" (at %p).\n",
-                             x.c_str(), (wchar_t*)*ptr, ptr);
-            break;
-        }
-        (*ptr) += stringLen;
-    }
-    template <class T>
-    void DoClass(T& x) {
-        x.DoState(*this);
-    }
-    template <class T>
-    void DoClass(T*& x) {
-        if (mode == MODE_READ) {
-            if (x != nullptr)
-                delete x;
-            x = new T();
-        }
-        x->DoState(*this);
-    }
-    template <class T>
-    void DoArray(T* x, int count) {
-        DoHelper<T>::DoArray(this, x, count);
-    }
-    template <class T>
-    void Do(T& x) {
-        DoHelper<T>::Do(this, x);
-    }
-    template <class T>
-    void DoPOD(T& x) {
-        DoHelper<T>::Do(this, x);
-    }
-    template <class T>
-    void DoPointer(T*& x, T* const base) {
-        // pointers can be more than 2^31 apart, but you're using this function wrong if you need
-        // that much range
-        s32 offset = x - base;
-        Do(offset);
-        if (mode == MODE_READ)
-            x = base + offset;
-    }
-    template <class T, LinkedListItem<T>* (*TNew)(), void (*TFree)(LinkedListItem<T>*),
-              void (*TDo)(PointerWrap&, T*)>
-    void DoLinkedList(LinkedListItem<T>*& list_start, LinkedListItem<T>** list_end = nullptr) {
-        LinkedListItem<T>* list_cur = list_start;
-        LinkedListItem<T>* prev = nullptr;
-        while (true) {
-            u8 shouldExist = (list_cur ? 1 : 0);
-            Do(shouldExist);
-            if (shouldExist == 1) {
-                LinkedListItem<T>* cur = list_cur ? list_cur : TNew();
-                TDo(*this, (T*)cur);
-                if (!list_cur) {
-                    if (mode == MODE_READ) {
-                        cur->next = nullptr;
-                        list_cur = cur;
-                        if (prev)
-                            prev->next = cur;
-                        else
-                            list_start = cur;
-                    } else {
-                        TFree(cur);
-                        continue;
-                    }
-                }
-            } else {
-                if (mode == MODE_READ) {
-                    if (prev)
-                        prev->next = nullptr;
-                    if (list_end)
-                        *list_end = prev;
-                    if (list_cur) {
-                        if (list_start == list_cur)
-                            list_start = nullptr;
-                        do {
-                            LinkedListItem<T>* next = list_cur->next;
-                            TFree(list_cur);
-                            list_cur = next;
-                        } while (list_cur);
-                    }
-                }
-                break;
-            }
-            prev = list_cur;
-            list_cur = list_cur->next;
-        }
-    }
-    void DoMarker(const char* prevName, u32 arbitraryNumber = 0x42) {
-        u32 cookie = arbitraryNumber;
-        Do(cookie);
-        if (mode == PointerWrap::MODE_READ && cookie != arbitraryNumber) {
-            LOG_ERROR(Common,
-                      "After \"%s\", found %d (0x%X) instead of save marker %d (0x%X). "
-                      "Aborting savestate load...",
-                      prevName, cookie, cookie, arbitraryNumber, arbitraryNumber);
-            SetError(ERROR_FAILURE);
-        }
-    }
-};
-inline PointerWrapSection::~PointerWrapSection() {
-    if (ver_ > 0) {
-        p_.DoMarker(title_);
-    }
-}
diff --git a/src/common/linear_disk_cache.h b/src/common/linear_disk_cache.h
deleted file mode 100644
index 94c695163..000000000
--- a/src/common/linear_disk_cache.h
+++ /dev/null
@@ -1,167 +0,0 @@
-// Copyright 2013 Dolphin Emulator Project / 2014 Citra Emulator Project
-// Licensed under GPLv2 or any later version
-// Refer to the license.txt file included.
-#pragma once
-#include <fstream>
-#include "common/common_types.h"
-// defined in Version.cpp
-extern const char* scm_rev_git_str;
-// On disk format:
-// header{
-// u32 'DCAC';
-// u32 version;  // svn_rev
-// u16 sizeof(key_type);
-// u16 sizeof(value_type);
-//}
-// key_value_pair{
-// u32 value_size;
-// key_type   key;
-// value_type[value_size]   value;
-//}
-template <typename K, typename V>
-class LinearDiskCacheReader {
-public:
-    virtual void Read(const K& key, const V* value, u32 value_size) = 0;
-};
-// Dead simple unsorted key-value store with append functionality.
-// No random read functionality, all reading is done in OpenAndRead.
-// Keys and values can contain any characters, including \0.
-//
-// Suitable for caching generated shader bytecode between executions.
-// Not tuned for extreme performance but should be reasonably fast.
-// Does not support keys or values larger than 2GB, which should be reasonable.
-// Keys must have non-zero length; values can have zero length.
-// K and V are some POD type
-// K : the key type
-// V : value array type
-template <typename K, typename V>
-class LinearDiskCache {
-public:
-    // return number of read entries
-    u32 OpenAndRead(const char* filename, LinearDiskCacheReader<K, V>& reader) {
-        using std::ios_base;
-        // close any currently opened file
-        Close();
-        m_num_entries = 0;
-        // try opening for reading/writing
-        OpenFStream(m_file, filename, ios_base::in | ios_base::out | ios_base::binary);
-        m_file.seekg(0, std::ios::end);
-        std::fstream::pos_type end_pos = m_file.tellg();
-        m_file.seekg(0, std::ios::beg);
-        std::fstream::pos_type start_pos = m_file.tellg();
-        std::streamoff file_size = end_pos - start_pos;
-        if (m_file.is_open() && ValidateHeader()) {
-            // good header, read some key/value pairs
-            K key;
-            V* value = nullptr;
-            u32 value_size;
-            u32 entry_number;
-            std::fstream::pos_type last_pos = m_file.tellg();
-            while (Read(&value_size)) {
-                std::streamoff next_extent =
-                    (last_pos - start_pos) + sizeof(value_size) + value_size;
-                if (next_extent > file_size)
-                    break;
-                delete[] value;
-                value = new V[value_size];
-                // read key/value and pass to reader
-                if (Read(&key) && Read(value, value_size) && Read(&entry_number) &&
-                    entry_number == m_num_entries + 1) {
-                    reader.Read(key, value, value_size);
-                } else {
-                    break;
-                }
-                m_num_entries++;
-                last_pos = m_file.tellg();
-            }
-            m_file.seekp(last_pos);
-            m_file.clear();
-            delete[] value;
-            return m_num_entries;
-        }
-        // failed to open file for reading or bad header
-        // close and recreate file
-        Close();
-        m_file.open(filename, ios_base::out | ios_base::trunc | ios_base::binary);
-        WriteHeader();
-        return 0;
-    }
-    void Sync() {
-        m_file.flush();
-    }
-    void Close() {
-        if (m_file.is_open())
-            m_file.close();
-        // clear any error flags
-        m_file.clear();
-    }
-    // Appends a key-value pair to the store.
-    void Append(const K& key, const V* value, u32 value_size) {
-        // TODO: Should do a check that we don't already have "key"? (I think each caller does that
-        // already.)
-        Write(&value_size);
-        Write(&key);
-        Write(value, value_size);
-        m_num_entries++;
-        Write(&m_num_entries);
-    }
-private:
-    void WriteHeader() {
-        Write(&m_header);
-    }
-    bool ValidateHeader() {
-        char file_header[sizeof(Header)];
-        return (Read(file_header, sizeof(Header)) &&
-                !memcmp((const char*)&m_header, file_header, sizeof(Header)));
-    }
-    template <typename D>
-    bool Write(const D* data, u32 count = 1) {
-        return m_file.write((const char*)data, count * sizeof(D)).good();
-    }
-    template <typename D>
-    bool Read(const D* data, u32 count = 1) {
-        return m_file.read((char*)data, count * sizeof(D)).good();
-    }
-    struct Header {
-        Header() : id(*(u32*)"DCAC"), key_t_size(sizeof(K)), value_t_size(sizeof(V)) {
-            memcpy(ver, scm_rev_git_str, 40);
-        }
-        const u32 id;
-        const u16 key_t_size, value_t_size;
-        char ver[40];
-    } m_header;
-    std::fstream m_file;
-    u32 m_num_entries;
-};
diff --git a/src/core/CMakeLists.txt b/src/core/CMakeLists.txt
index c2a6f56cd..45dbd6796 100644
--- a/src/core/CMakeLists.txt
+++ b/src/core/CMakeLists.txt
@@ -42,8 +42,6 @@ add_library(core STATIC
    hle/kernel/client_port.h
    hle/kernel/client_session.cpp
    hle/kernel/client_session.h
-    hle/kernel/condition_variable.cpp
-    hle/kernel/condition_variable.h
    hle/kernel/errors.h
    hle/kernel/event.cpp
    hle/kernel/event.h
@@ -183,10 +181,10 @@ add_library(core STATIC
    hle/service/nvflinger/buffer_queue.h
    hle/service/nvflinger/nvflinger.cpp
    hle/service/nvflinger/nvflinger.h
+    hle/service/pctl/module.cpp
+    hle/service/pctl/module.h
    hle/service/pctl/pctl.cpp
    hle/service/pctl/pctl.h
-    hle/service/pctl/pctl_a.cpp
-    hle/service/pctl/pctl_a.h
    hle/service/prepo/prepo.cpp
    hle/service/prepo/prepo.h
    hle/service/service.cpp
diff --git a/src/core/file_sys/disk_filesystem.cpp b/src/core/file_sys/disk_filesystem.cpp
index ca1323873..8aa0e0aa4 100644
--- a/src/core/file_sys/disk_filesystem.cpp
+++ b/src/core/file_sys/disk_filesystem.cpp
@@ -67,26 +67,32 @@ ResultCode Disk_FileSystem::DeleteFile(const std::string& path) const {
    return RESULT_SUCCESS;
 }
-ResultCode Disk_FileSystem::RenameFile(const Path& src_path, const Path& dest_path) const {
+ResultCode Disk_FileSystem::RenameFile(const std::string& src_path,
-    LOG_WARNING(Service_FS, "(STUBBED) called");
+                                       const std::string& dest_path) const {
+    const std::string full_src_path = base_directory + src_path;
+    const std::string full_dest_path = base_directory + dest_path;
+    if (!FileUtil::Exists(full_src_path)) {
+        return ERROR_PATH_NOT_FOUND;
+    }
    // TODO(wwylele): Use correct error code
-    return ResultCode(-1);
+    return FileUtil::Rename(full_src_path, full_dest_path) ? RESULT_SUCCESS : ResultCode(-1);
 }
 ResultCode Disk_FileSystem::DeleteDirectory(const Path& path) const {
-    LOG_WARNING(Service_FS, "(STUBBED) called");
+    NGLOG_WARNING(Service_FS, "(STUBBED) called");
    // TODO(wwylele): Use correct error code
    return ResultCode(-1);
 }
 ResultCode Disk_FileSystem::DeleteDirectoryRecursively(const Path& path) const {
-    LOG_WARNING(Service_FS, "(STUBBED) called");
+    NGLOG_WARNING(Service_FS, "(STUBBED) called");
    // TODO(wwylele): Use correct error code
    return ResultCode(-1);
 }
 ResultCode Disk_FileSystem::CreateFile(const std::string& path, u64 size) const {
-    LOG_WARNING(Service_FS, "(STUBBED) called");
+    NGLOG_WARNING(Service_FS, "(STUBBED) called");
    std::string full_path = base_directory + path;
    if (size == 0) {
@@ -101,7 +107,7 @@ ResultCode Disk_FileSystem::CreateFile(const std::string& path, u64 size) const
        return RESULT_SUCCESS;
    }
-    LOG_ERROR(Service_FS, "Too large file");
+    NGLOG_ERROR(Service_FS, "Too large file");
    // TODO(Subv): Find out the correct error code
    return ResultCode(-1);
 }
@@ -114,13 +120,13 @@ ResultCode Disk_FileSystem::CreateDirectory(const std::string& path) const {
        return RESULT_SUCCESS;
    }
-    LOG_CRITICAL(Service_FS, "(unreachable) Unknown error creating %s", full_path.c_str());
+    NGLOG_CRITICAL(Service_FS, "(unreachable) Unknown error creating {}", full_path);
    // TODO(wwylele): Use correct error code
    return ResultCode(-1);
 }
 ResultCode Disk_FileSystem::RenameDirectory(const Path& src_path, const Path& dest_path) const {
-    LOG_WARNING(Service_FS, "(STUBBED) called");
+    NGLOG_WARNING(Service_FS, "(STUBBED) called");
    // TODO(wwylele): Use correct error code
    return ResultCode(-1);
 }
@@ -140,7 +146,7 @@ ResultVal<std::unique_ptr<DirectoryBackend>> Disk_FileSystem::OpenDirectory(
 }
 u64 Disk_FileSystem::GetFreeSpaceSize() const {
-    LOG_WARNING(Service_FS, "(STUBBED) called");
+    NGLOG_WARNING(Service_FS, "(STUBBED) called");
    return 0;
 }
@@ -157,14 +163,14 @@ ResultVal<FileSys::EntryType> Disk_FileSystem::GetEntryType(const std::string& p
 }
 ResultVal<size_t> Disk_Storage::Read(const u64 offset, const size_t length, u8* buffer) const {
-    LOG_TRACE(Service_FS, "called offset=%llu, length=%zu", offset, length);
+    NGLOG_TRACE(Service_FS, "called offset={}, length={}", offset, length);
    file->Seek(offset, SEEK_SET);
    return MakeResult<size_t>(file->ReadBytes(buffer, length));
 }
 ResultVal<size_t> Disk_Storage::Write(const u64 offset, const size_t length, const bool flush,
                                      const u8* buffer) const {
-    LOG_WARNING(Service_FS, "(STUBBED) called");
+    NGLOG_WARNING(Service_FS, "(STUBBED) called");
    file->Seek(offset, SEEK_SET);
    size_t written = file->WriteBytes(buffer, length);
    if (flush) {
@@ -198,8 +204,7 @@ u64 Disk_Directory::Read(const u64 count, Entry* entries) {
        const std::string& filename = file.virtualName;
        Entry& entry = entries[entries_read];
-        LOG_TRACE(Service_FS, "File %s: size=%llu dir=%d", filename.c_str(), file.size,
+        NGLOG_TRACE(Service_FS, "File {}: size={} dir={}", filename, file.size, file.isDirectory);
-                  file.isDirectory);
        // TODO(Link Mauve): use a proper conversion to UTF-16.
        for (size_t j = 0; j < FILENAME_LENGTH; ++j) {
diff --git a/src/core/file_sys/disk_filesystem.h b/src/core/file_sys/disk_filesystem.h
index 8f9e1145a..591e39fda 100644
--- a/src/core/file_sys/disk_filesystem.h
+++ b/src/core/file_sys/disk_filesystem.h
@@ -26,7 +26,7 @@ public:
    ResultVal<std::unique_ptr<StorageBackend>> OpenFile(const std::string& path,
                                                        Mode mode) const override;
    ResultCode DeleteFile(const std::string& path) const override;
-    ResultCode RenameFile(const Path& src_path, const Path& dest_path) const override;
+    ResultCode RenameFile(const std::string& src_path, const std::string& dest_path) const override;
    ResultCode DeleteDirectory(const Path& path) const override;
    ResultCode DeleteDirectoryRecursively(const Path& path) const override;
    ResultCode CreateFile(const std::string& path, u64 size) const override;
diff --git a/src/core/file_sys/filesystem.cpp b/src/core/file_sys/filesystem.cpp
index 82fdb3c46..87083878b 100644
--- a/src/core/file_sys/filesystem.cpp
+++ b/src/core/file_sys/filesystem.cpp
@@ -71,7 +71,7 @@ std::string Path::AsString() const {
    case Binary:
    default:
        // TODO(yuriks): Add assert
-        LOG_ERROR(Service_FS, "LowPathType cannot be converted to string!");
+        NGLOG_ERROR(Service_FS, "LowPathType cannot be converted to string!");
        return {};
    }
 }
@@ -87,7 +87,7 @@ std::u16string Path::AsU16Str() const {
    case Invalid:
    case Binary:
        // TODO(yuriks): Add assert
-        LOG_ERROR(Service_FS, "LowPathType cannot be converted to u16string!");
+        NGLOG_ERROR(Service_FS, "LowPathType cannot be converted to u16string!");
        return {};
    }
@@ -115,7 +115,7 @@ std::vector<u8> Path::AsBinary() const {
    case Invalid:
    default:
        // TODO(yuriks): Add assert
-        LOG_ERROR(Service_FS, "LowPathType cannot be converted to binary!");
+        NGLOG_ERROR(Service_FS, "LowPathType cannot be converted to binary!");
        return {};
    }
 }
diff --git a/src/core/file_sys/filesystem.h b/src/core/file_sys/filesystem.h
index beefcfdb2..295a3133e 100644
--- a/src/core/file_sys/filesystem.h
+++ b/src/core/file_sys/filesystem.h
@@ -126,7 +126,8 @@ public:
     * @param dest_path Destination path relative to the archive
     * @return Result of the operation
     */
-    virtual ResultCode RenameFile(const Path& src_path, const Path& dest_path) const = 0;
+    virtual ResultCode RenameFile(const std::string& src_path,
+                                  const std::string& dest_path) const = 0;
    /**
     * Rename a Directory specified by its path
diff --git a/src/core/file_sys/partition_filesystem.cpp b/src/core/file_sys/partition_filesystem.cpp
index 4a58a9291..808254ecc 100644
--- a/src/core/file_sys/partition_filesystem.cpp
+++ b/src/core/file_sys/partition_filesystem.cpp
@@ -2,7 +2,6 @@
 // Licensed under GPLv2 or any later version
 // Refer to the license.txt file included.
-#include <cinttypes>
 #include <utility>
 #include "common/file_util.h"
 #include "common/logging/log.h"
@@ -40,7 +39,7 @@ Loader::ResultStatus PartitionFilesystem::Load(const std::string& file_path, siz
    Loader::ResultStatus result = Load(file_data);
    if (result != Loader::ResultStatus::Success)
-        LOG_ERROR(Service_FS, "Failed to load PFS from file %s!", file_path.c_str());
+        NGLOG_ERROR(Service_FS, "Failed to load PFS from file {}!", file_path);
    return result;
 }
diff --git a/src/core/file_sys/program_metadata.cpp b/src/core/file_sys/program_metadata.cpp
index a6dcebcc3..1f5ded514 100644
--- a/src/core/file_sys/program_metadata.cpp
+++ b/src/core/file_sys/program_metadata.cpp
@@ -2,7 +2,6 @@
 // Licensed under GPLv2 or any later version
 // Refer to the license.txt file included.
-#include <cinttypes>
 #include "common/file_util.h"
 #include "common/logging/log.h"
 #include "core/file_sys/program_metadata.h"
@@ -22,7 +21,7 @@ Loader::ResultStatus ProgramMetadata::Load(const std::string& file_path) {
    Loader::ResultStatus result = Load(file_data);
    if (result != Loader::ResultStatus::Success)
-        LOG_ERROR(Service_FS, "Failed to load NPDM from file %s!", file_path.c_str());
+        NGLOG_ERROR(Service_FS, "Failed to load NPDM from file {}!", file_path);
    return result;
 }
@@ -77,14 +76,14 @@ u64 ProgramMetadata::GetFilesystemPermissions() const {
 }
 void ProgramMetadata::Print() const {
-    LOG_DEBUG(Service_FS, "Magic:                  %.4s", npdm_header.magic.data());
+    NGLOG_DEBUG(Service_FS, "Magic:                  {:.4}", npdm_header.magic.data());
-    LOG_DEBUG(Service_FS, "Main thread priority:   0x%02x", npdm_header.main_thread_priority);
+    NGLOG_DEBUG(Service_FS, "Main thread priority:   {:#04X}", npdm_header.main_thread_priority);
-    LOG_DEBUG(Service_FS, "Main thread core:       %u", npdm_header.main_thread_cpu);
+    NGLOG_DEBUG(Service_FS, "Main thread core:       {}", npdm_header.main_thread_cpu);
-    LOG_DEBUG(Service_FS, "Main thread stack size: 0x%x bytes", npdm_header.main_stack_size);
+    NGLOG_DEBUG(Service_FS, "Main thread stack size: {:#X} bytes", npdm_header.main_stack_size);
-    LOG_DEBUG(Service_FS, "Process category:       %u", npdm_header.process_category);
+    NGLOG_DEBUG(Service_FS, "Process category:       {}", npdm_header.process_category);
-    LOG_DEBUG(Service_FS, "Flags:                  %02x", npdm_header.flags);
+    NGLOG_DEBUG(Service_FS, "Flags:                  {:02X}", npdm_header.flags);
-    LOG_DEBUG(Service_FS, " > 64-bit instructions: %s",
+    NGLOG_DEBUG(Service_FS, " > 64-bit instructions: {}",
-              npdm_header.has_64_bit_instructions ? "YES" : "NO");
+                npdm_header.has_64_bit_instructions ? "YES" : "NO");
    auto address_space = "Unknown";
    switch (npdm_header.address_space_type) {
@@ -96,19 +95,19 @@ void ProgramMetadata::Print() const {
        break;
    }
-    LOG_DEBUG(Service_FS, " > Address space:       %s\n", address_space);
+    NGLOG_DEBUG(Service_FS, " > Address space:       {}\n", address_space);
    // Begin ACID printing (potential perms, signed)
-    LOG_DEBUG(Service_FS, "Magic:                  %.4s", acid_header.magic.data());
+    NGLOG_DEBUG(Service_FS, "Magic:                  {:.4}", acid_header.magic.data());
-    LOG_DEBUG(Service_FS, "Flags:                  %02x", acid_header.flags);
+    NGLOG_DEBUG(Service_FS, "Flags:                  {:02X}", acid_header.flags);
-    LOG_DEBUG(Service_FS, " > Is Retail:           %s", acid_header.is_retail ? "YES" : "NO");
+    NGLOG_DEBUG(Service_FS, " > Is Retail:           {}", acid_header.is_retail ? "YES" : "NO");
-    LOG_DEBUG(Service_FS, "Title ID Min:           %016" PRIX64, acid_header.title_id_min);
+    NGLOG_DEBUG(Service_FS, "Title ID Min:           {:016X}", acid_header.title_id_min);
-    LOG_DEBUG(Service_FS, "Title ID Max:           %016" PRIX64, acid_header.title_id_max);
+    NGLOG_DEBUG(Service_FS, "Title ID Max:           {:016X}", acid_header.title_id_max);
-    LOG_DEBUG(Service_FS, "Filesystem Access:      %016" PRIX64 "\n", acid_file_access.permissions);
+    NGLOG_DEBUG(Service_FS, "Filesystem Access:      {:016X}\n", acid_file_access.permissions);
    // Begin ACI0 printing (actual perms, unsigned)
-    LOG_DEBUG(Service_FS, "Magic:                  %.4s", aci_header.magic.data());
+    NGLOG_DEBUG(Service_FS, "Magic:                  {:.4}", aci_header.magic.data());
-    LOG_DEBUG(Service_FS, "Title ID:               %016" PRIX64, aci_header.title_id);
+    NGLOG_DEBUG(Service_FS, "Title ID:               {:016X}", aci_header.title_id);
-    LOG_DEBUG(Service_FS, "Filesystem Access:      %016" PRIX64 "\n", aci_file_access.permissions);
+    NGLOG_DEBUG(Service_FS, "Filesystem Access:      {:016X}\n", aci_file_access.permissions);
 }
 } // namespace FileSys
diff --git a/src/core/file_sys/romfs_factory.cpp b/src/core/file_sys/romfs_factory.cpp
index b21427948..dc7591aca 100644
--- a/src/core/file_sys/romfs_factory.cpp
+++ b/src/core/file_sys/romfs_factory.cpp
@@ -14,7 +14,7 @@ namespace FileSys {
 RomFS_Factory::RomFS_Factory(Loader::AppLoader& app_loader) {
    // Load the RomFS from the app
    if (Loader::ResultStatus::Success != app_loader.ReadRomFS(romfs_file, data_offset, data_size)) {
-        LOG_ERROR(Service_FS, "Unable to read RomFS!");
+        NGLOG_ERROR(Service_FS, "Unable to read RomFS!");
    }
 }
@@ -24,13 +24,13 @@ ResultVal<std::unique_ptr<FileSystemBackend>> RomFS_Factory::Open(const Path& pa
 }
 ResultCode RomFS_Factory::Format(const Path& path) {
-    LOG_ERROR(Service_FS, "Unimplemented Format archive %s", GetName().c_str());
+    NGLOG_ERROR(Service_FS, "Unimplemented Format archive {}", GetName());
    // TODO(bunnei): Find the right error code for this
    return ResultCode(-1);
 }
 ResultVal<ArchiveFormatInfo> RomFS_Factory::GetFormatInfo(const Path& path) const {
-    LOG_ERROR(Service_FS, "Unimplemented GetFormatInfo archive %s", GetName().c_str());
+    NGLOG_ERROR(Service_FS, "Unimplemented GetFormatInfo archive {}", GetName());
    // TODO(bunnei): Find the right error code for this
    return ResultCode(-1);
 }
diff --git a/src/core/file_sys/romfs_filesystem.cpp b/src/core/file_sys/romfs_filesystem.cpp
index 3d77e2d5f..8e2bce687 100644
--- a/src/core/file_sys/romfs_filesystem.cpp
+++ b/src/core/file_sys/romfs_filesystem.cpp
@@ -21,73 +21,72 @@ ResultVal<std::unique_ptr<StorageBackend>> RomFS_FileSystem::OpenFile(const std:
 }
 ResultCode RomFS_FileSystem::DeleteFile(const std::string& path) const {
-    LOG_CRITICAL(Service_FS, "Attempted to delete a file from an ROMFS archive (%s).",
+    NGLOG_CRITICAL(Service_FS, "Attempted to delete a file from an ROMFS archive ({}).", GetName());
-                 GetName().c_str());
    // TODO(bunnei): Use correct error code
    return ResultCode(-1);
 }
-ResultCode RomFS_FileSystem::RenameFile(const Path& src_path, const Path& dest_path) const {
+ResultCode RomFS_FileSystem::RenameFile(const std::string& src_path,
-    LOG_CRITICAL(Service_FS, "Attempted to rename a file within an ROMFS archive (%s).",
+                                        const std::string& dest_path) const {
-                 GetName().c_str());
+    NGLOG_CRITICAL(Service_FS, "Attempted to rename a file within an ROMFS archive ({}).",
+                   GetName());
    // TODO(wwylele): Use correct error code
    return ResultCode(-1);
 }
 ResultCode RomFS_FileSystem::DeleteDirectory(const Path& path) const {
-    LOG_CRITICAL(Service_FS, "Attempted to delete a directory from an ROMFS archive (%s).",
+    NGLOG_CRITICAL(Service_FS, "Attempted to delete a directory from an ROMFS archive ({}).",
-                 GetName().c_str());
+                   GetName());
    // TODO(wwylele): Use correct error code
    return ResultCode(-1);
 }
 ResultCode RomFS_FileSystem::DeleteDirectoryRecursively(const Path& path) const {
-    LOG_CRITICAL(Service_FS, "Attempted to delete a directory from an ROMFS archive (%s).",
+    NGLOG_CRITICAL(Service_FS, "Attempted to delete a directory from an ROMFS archive ({}).",
-                 GetName().c_str());
+                   GetName());
    // TODO(wwylele): Use correct error code
    return ResultCode(-1);
 }
 ResultCode RomFS_FileSystem::CreateFile(const std::string& path, u64 size) const {
-    LOG_CRITICAL(Service_FS, "Attempted to create a file in an ROMFS archive (%s).",
+    NGLOG_CRITICAL(Service_FS, "Attempted to create a file in an ROMFS archive ({}).", GetName());
-                 GetName().c_str());
    // TODO(bunnei): Use correct error code
    return ResultCode(-1);
 }
 ResultCode RomFS_FileSystem::CreateDirectory(const std::string& path) const {
-    LOG_CRITICAL(Service_FS, "Attempted to create a directory in an ROMFS archive (%s).",
+    NGLOG_CRITICAL(Service_FS, "Attempted to create a directory in an ROMFS archive ({}).",
-                 GetName().c_str());
+                   GetName());
    // TODO(wwylele): Use correct error code
    return ResultCode(-1);
 }
 ResultCode RomFS_FileSystem::RenameDirectory(const Path& src_path, const Path& dest_path) const {
-    LOG_CRITICAL(Service_FS, "Attempted to rename a file within an ROMFS archive (%s).",
+    NGLOG_CRITICAL(Service_FS, "Attempted to rename a file within an ROMFS archive ({}).",
-                 GetName().c_str());
+                   GetName());
    // TODO(wwylele): Use correct error code
    return ResultCode(-1);
 }
 ResultVal<std::unique_ptr<DirectoryBackend>> RomFS_FileSystem::OpenDirectory(
    const std::string& path) const {
-    LOG_WARNING(Service_FS, "Opening Directory in a ROMFS archive");
+    NGLOG_WARNING(Service_FS, "Opening Directory in a ROMFS archive");
    return MakeResult<std::unique_ptr<DirectoryBackend>>(std::make_unique<ROMFSDirectory>());
 }
 u64 RomFS_FileSystem::GetFreeSpaceSize() const {
-    LOG_WARNING(Service_FS, "Attempted to get the free space in an ROMFS archive");
+    NGLOG_WARNING(Service_FS, "Attempted to get the free space in an ROMFS archive");
    return 0;
 }
 ResultVal<FileSys::EntryType> RomFS_FileSystem::GetEntryType(const std::string& path) const {
-    LOG_CRITICAL(Service_FS, "Called within an ROMFS archive (path %s).", path.c_str());
+    NGLOG_CRITICAL(Service_FS, "Called within an ROMFS archive (path {}).", path);
    // TODO(wwylele): Use correct error code
    return ResultCode(-1);
 }
 ResultVal<size_t> RomFS_Storage::Read(const u64 offset, const size_t length, u8* buffer) const {
-    LOG_TRACE(Service_FS, "called offset=%llu, length=%zu", offset, length);
+    NGLOG_TRACE(Service_FS, "called offset={}, length={}", offset, length);
    romfs_file->Seek(data_offset + offset, SEEK_SET);
    size_t read_length = (size_t)std::min((u64)length, data_size - offset);
@@ -96,7 +95,7 @@ ResultVal<size_t> RomFS_Storage::Read(const u64 offset, const size_t length, u8*
 ResultVal<size_t> RomFS_Storage::Write(const u64 offset, const size_t length, const bool flush,
                                       const u8* buffer) const {
-    LOG_ERROR(Service_FS, "Attempted to write to ROMFS file");
+    NGLOG_ERROR(Service_FS, "Attempted to write to ROMFS file");
    // TODO(Subv): Find error code
    return MakeResult<size_t>(0);
 }
@@ -106,7 +105,7 @@ u64 RomFS_Storage::GetSize() const {
 }
 bool RomFS_Storage::SetSize(const u64 size) const {
-    LOG_ERROR(Service_FS, "Attempted to set the size of an ROMFS file");
+    NGLOG_ERROR(Service_FS, "Attempted to set the size of an ROMFS file");
    return false;
 }
diff --git a/src/core/file_sys/romfs_filesystem.h b/src/core/file_sys/romfs_filesystem.h
index 1b5cac409..ba9d85823 100644
--- a/src/core/file_sys/romfs_filesystem.h
+++ b/src/core/file_sys/romfs_filesystem.h
@@ -32,7 +32,7 @@ public:
    ResultVal<std::unique_ptr<StorageBackend>> OpenFile(const std::string& path,
                                                        Mode mode) const override;
    ResultCode DeleteFile(const std::string& path) const override;
-    ResultCode RenameFile(const Path& src_path, const Path& dest_path) const override;
+    ResultCode RenameFile(const std::string& src_path, const std::string& dest_path) const override;
    ResultCode DeleteDirectory(const Path& path) const override;
    ResultCode DeleteDirectoryRecursively(const Path& path) const override;
    ResultCode CreateFile(const std::string& path, u64 size) const override;
diff --git a/src/core/file_sys/savedata_factory.cpp b/src/core/file_sys/savedata_factory.cpp
index 14868fed2..c1be8fee4 100644
--- a/src/core/file_sys/savedata_factory.cpp
+++ b/src/core/file_sys/savedata_factory.cpp
@@ -2,11 +2,9 @@
 // Licensed under GPLv2 or any later version
 // Refer to the license.txt file included.
-#include <cinttypes>
 #include <memory>
 #include "common/common_types.h"
 #include "common/logging/log.h"
-#include "common/string_util.h"
 #include "core/core.h"
 #include "core/file_sys/disk_filesystem.h"
 #include "core/file_sys/savedata_factory.h"
@@ -30,7 +28,7 @@ ResultVal<std::unique_ptr<FileSystemBackend>> SaveData_Factory::Open(const Path&
 }
 ResultCode SaveData_Factory::Format(const Path& path) {
-    LOG_WARNING(Service_FS, "Format archive %s", GetName().c_str());
+    NGLOG_WARNING(Service_FS, "Format archive {}", GetName());
    // Create the save data directory.
    if (!FileUtil::CreateFullPath(GetFullPath())) {
        // TODO(Subv): Find the correct error code.
@@ -41,7 +39,7 @@ ResultCode SaveData_Factory::Format(const Path& path) {
 }
 ResultVal<ArchiveFormatInfo> SaveData_Factory::GetFormatInfo(const Path& path) const {
-    LOG_ERROR(Service_FS, "Unimplemented GetFormatInfo archive %s", GetName().c_str());
+    NGLOG_ERROR(Service_FS, "Unimplemented GetFormatInfo archive {}", GetName());
    // TODO(bunnei): Find the right error code for this
    return ResultCode(-1);
 }
@@ -50,8 +48,7 @@ std::string SaveData_Factory::GetFullPath() const {
    u64 title_id = Core::CurrentProcess()->program_id;
    // TODO(Subv): Somehow obtain this value.
    u32 user = 0;
-    return Common::StringFromFormat("%ssave/%016" PRIX64 "/%08X/", nand_directory.c_str(), title_id,
+    return fmt::format("{}save/{:016X}/{:08X}/", nand_directory, title_id, user);
-                                    user);
 }
 } // namespace FileSys
diff --git a/src/core/file_sys/sdmc_factory.cpp b/src/core/file_sys/sdmc_factory.cpp
index 00e80d2a7..59ac3e0be 100644
--- a/src/core/file_sys/sdmc_factory.cpp
+++ b/src/core/file_sys/sdmc_factory.cpp
@@ -2,7 +2,6 @@
 // Licensed under GPLv2 or any later version
 // Refer to the license.txt file included.
-#include <cinttypes>
 #include <memory>
 #include "common/common_types.h"
 #include "common/logging/log.h"
@@ -26,13 +25,13 @@ ResultVal<std::unique_ptr<FileSystemBackend>> SDMC_Factory::Open(const Path& pat
 }
 ResultCode SDMC_Factory::Format(const Path& path) {
-    LOG_ERROR(Service_FS, "Unimplemented Format archive %s", GetName().c_str());
+    NGLOG_ERROR(Service_FS, "Unimplemented Format archive {}", GetName());
    // TODO(Subv): Find the right error code for this
    return ResultCode(-1);
 }
 ResultVal<ArchiveFormatInfo> SDMC_Factory::GetFormatInfo(const Path& path) const {
-    LOG_ERROR(Service_FS, "Unimplemented GetFormatInfo archive %s", GetName().c_str());
+    NGLOG_ERROR(Service_FS, "Unimplemented GetFormatInfo archive {}", GetName());
    // TODO(bunnei): Find the right error code for this
    return ResultCode(-1);
 }
diff --git a/src/core/gdbstub/gdbstub.cpp b/src/core/gdbstub/gdbstub.cpp
index e4f337a0a..46606b992 100644
--- a/src/core/gdbstub/gdbstub.cpp
+++ b/src/core/gdbstub/gdbstub.cpp
@@ -6,7 +6,6 @@
 #include <algorithm>
 #include <atomic>
-#include <cinttypes>
 #include <climits>
 #include <csignal>
 #include <cstdarg>
@@ -180,7 +179,7 @@ static u8 HexCharToValue(u8 hex) {
        return hex - 'A' + 0xA;
    }
-    LOG_ERROR(Debug_GDBStub, "Invalid nibble: %c (%02x)\n", hex, hex);
+    NGLOG_ERROR(Debug_GDBStub, "Invalid nibble: {} ({:02X})", hex, hex);
    return 0;
 }
@@ -320,7 +319,7 @@ static u8 ReadByte() {
    u8 c;
    size_t received_size = recv(gdbserver_socket, reinterpret_cast<char*>(&c), 1, MSG_WAITALL);
    if (received_size != 1) {
-        LOG_ERROR(Debug_GDBStub, "recv failed : %ld", received_size);
+        NGLOG_ERROR(Debug_GDBStub, "recv failed: {}", received_size);
        Shutdown();
    }
@@ -361,9 +360,8 @@ static void RemoveBreakpoint(BreakpointType type, PAddr addr) {
    auto bp = p.find(static_cast<u64>(addr));
    if (bp != p.end()) {
-        LOG_DEBUG(Debug_GDBStub,
+        NGLOG_DEBUG(Debug_GDBStub, "gdb: removed a breakpoint: {:016X} bytes at {:016X} of type {}",
-                  "gdb: removed a breakpoint: %016" PRIx64 " bytes at %016" PRIx64 " of type %d\n",
+                    bp->second.len, bp->second.addr, static_cast<int>(type));
-                  bp->second.len, bp->second.addr, static_cast<int>(type));
        p.erase(static_cast<u64>(addr));
    }
 }
@@ -408,10 +406,10 @@ bool CheckBreakpoint(PAddr addr, BreakpointType type) {
        }
        if (bp->second.active && (addr >= bp->second.addr && addr < bp->second.addr + len)) {
-            LOG_DEBUG(Debug_GDBStub,
+            NGLOG_DEBUG(Debug_GDBStub,
-                      "Found breakpoint type %d @ %016" PRIx64 ", range: %016" PRIx64
+                        "Found breakpoint type {} @ {:016X}, range: {:016X}"
-                      " - %016" PRIx64 " (%" PRIx64 " bytes)\n",
+                        " - {:016X} ({:X} bytes)",
-                      static_cast<int>(type), addr, bp->second.addr, bp->second.addr + len, len);
+                        static_cast<int>(type), addr, bp->second.addr, bp->second.addr + len, len);
            return true;
        }
    }
@@ -427,7 +425,7 @@ bool CheckBreakpoint(PAddr addr, BreakpointType type) {
 static void SendPacket(const char packet) {
    size_t sent_size = send(gdbserver_socket, &packet, 1, 0);
    if (sent_size != 1) {
-        LOG_ERROR(Debug_GDBStub, "send failed");
+        NGLOG_ERROR(Debug_GDBStub, "send failed");
    }
 }
@@ -445,7 +443,7 @@ static void SendReply(const char* reply) {
    command_length = static_cast<u32>(strlen(reply));
    if (command_length + 4 > sizeof(command_buffer)) {
-        LOG_ERROR(Debug_GDBStub, "command_buffer overflow in SendReply");
+        NGLOG_ERROR(Debug_GDBStub, "command_buffer overflow in SendReply");
        return;
    }
@@ -462,7 +460,7 @@ static void SendReply(const char* reply) {
    while (left > 0) {
        int sent_size = send(gdbserver_socket, reinterpret_cast<char*>(ptr), left, 0);
        if (sent_size < 0) {
-            LOG_ERROR(Debug_GDBStub, "gdb: send failed");
+            NGLOG_ERROR(Debug_GDBStub, "gdb: send failed");
            return Shutdown();
        }
@@ -473,7 +471,7 @@ static void SendReply(const char* reply) {
 /// Handle query command from gdb client.
 static void HandleQuery() {
-    LOG_DEBUG(Debug_GDBStub, "gdb: query '%s'\n", command_buffer + 1);
+    NGLOG_DEBUG(Debug_GDBStub, "gdb: query '{}'", command_buffer + 1);
    const char* query = reinterpret_cast<const char*>(command_buffer + 1);
@@ -512,8 +510,8 @@ static void SendSignal(u32 signal) {
    latest_signal = signal;
-    std::string buffer = Common::StringFromFormat("T%02x", latest_signal);
+    std::string buffer = fmt::format("T{:02x}", latest_signal);
-    LOG_DEBUG(Debug_GDBStub, "Response: %s", buffer.c_str());
+    NGLOG_DEBUG(Debug_GDBStub, "Response: {}", buffer);
    SendReply(buffer.c_str());
 }
@@ -527,18 +525,18 @@ static void ReadCommand() {
        // ignore ack
        return;
    } else if (c == 0x03) {
-        LOG_INFO(Debug_GDBStub, "gdb: found break command\n");
+        NGLOG_INFO(Debug_GDBStub, "gdb: found break command");
        halt_loop = true;
        SendSignal(SIGTRAP);
        return;
    } else if (c != GDB_STUB_START) {
-        LOG_DEBUG(Debug_GDBStub, "gdb: read invalid byte %02x\n", c);
+        NGLOG_DEBUG(Debug_GDBStub, "gdb: read invalid byte {:02X}", c);
        return;
    }
    while ((c = ReadByte()) != GDB_STUB_END) {
        if (command_length >= sizeof(command_buffer)) {
-            LOG_ERROR(Debug_GDBStub, "gdb: command_buffer overflow\n");
+            NGLOG_ERROR(Debug_GDBStub, "gdb: command_buffer overflow");
            SendPacket(GDB_STUB_NACK);
            return;
        }
@@ -551,9 +549,10 @@ static void ReadCommand() {
    u8 checksum_calculated = CalculateChecksum(command_buffer, command_length);
    if (checksum_received != checksum_calculated) {
-        LOG_ERROR(Debug_GDBStub,
+        NGLOG_ERROR(
-                  "gdb: invalid checksum: calculated %02x and read %02x for $%s# (length: %d)\n",
+            Debug_GDBStub,
-                  checksum_calculated, checksum_received, command_buffer, command_length);
+            "gdb: invalid checksum: calculated {:02X} and read {:02X} for ${}# (length: {})",
+            checksum_calculated, checksum_received, command_buffer, command_length);
        command_length = 0;
@@ -580,7 +579,7 @@ static bool IsDataAvailable() {
    t.tv_usec = 0;
    if (select(gdbserver_socket + 1, &fd_socket, nullptr, nullptr, &t) < 0) {
-        LOG_ERROR(Debug_GDBStub, "select failed");
+        NGLOG_ERROR(Debug_GDBStub, "select failed");
        return false;
    }
@@ -693,7 +692,7 @@ static void ReadMemory() {
    u64 len =
        HexToLong(start_offset, static_cast<u64>((command_buffer + command_length) - start_offset));
-    LOG_DEBUG(Debug_GDBStub, "gdb: addr: %016lx len: %016lx\n", addr, len);
+    NGLOG_DEBUG(Debug_GDBStub, "gdb: addr: {:016X} len: {:016X}", addr, len);
    if (len * 2 > sizeof(reply)) {
        SendReply("E01");
@@ -781,8 +780,8 @@ static bool CommitBreakpoint(BreakpointType type, PAddr addr, u64 len) {
    breakpoint.len = len;
    p.insert({addr, breakpoint});
-    LOG_DEBUG(Debug_GDBStub, "gdb: added %d breakpoint: %016" PRIx64 " bytes at %016" PRIx64 "\n",
+    NGLOG_DEBUG(Debug_GDBStub, "gdb: added {} breakpoint: {:016X} bytes at {:016X}",
-              static_cast<int>(type), breakpoint.len, breakpoint.addr);
+                static_cast<int>(type), breakpoint.len, breakpoint.addr);
    return true;
 }
@@ -889,7 +888,7 @@ void HandlePacket() {
        return;
    }
-    LOG_DEBUG(Debug_GDBStub, "Packet: %s", command_buffer);
+    NGLOG_DEBUG(Debug_GDBStub, "Packet: {}", command_buffer);
    switch (command_buffer[0]) {
    case 'q':
@@ -903,7 +902,7 @@ void HandlePacket() {
        break;
    case 'k':
        Shutdown();
-        LOG_INFO(Debug_GDBStub, "killed by gdb");
+        NGLOG_INFO(Debug_GDBStub, "killed by gdb");
        return;
    case 'g':
        ReadRegisters();
@@ -982,7 +981,7 @@ static void Init(u16 port) {
    breakpoints_write.clear();
    // Start gdb server
-    LOG_INFO(Debug_GDBStub, "Starting GDB server on port %d...", port);
+    NGLOG_INFO(Debug_GDBStub, "Starting GDB server on port {}...", port);
    sockaddr_in saddr_server = {};
    saddr_server.sin_family = AF_INET;
@@ -995,28 +994,28 @@ static void Init(u16 port) {
    int tmpsock = static_cast<int>(socket(PF_INET, SOCK_STREAM, 0));
    if (tmpsock == -1) {
-        LOG_ERROR(Debug_GDBStub, "Failed to create gdb socket");
+        NGLOG_ERROR(Debug_GDBStub, "Failed to create gdb socket");
    }
    // Set socket to SO_REUSEADDR so it can always bind on the same port
    int reuse_enabled = 1;
    if (setsockopt(tmpsock, SOL_SOCKET, SO_REUSEADDR, (const char*)&reuse_enabled,
                   sizeof(reuse_enabled)) < 0) {
-        LOG_ERROR(Debug_GDBStub, "Failed to set gdb socket option");
+        NGLOG_ERROR(Debug_GDBStub, "Failed to set gdb socket option");
    }
    const sockaddr* server_addr = reinterpret_cast<const sockaddr*>(&saddr_server);
    socklen_t server_addrlen = sizeof(saddr_server);
    if (bind(tmpsock, server_addr, server_addrlen) < 0) {
-        LOG_ERROR(Debug_GDBStub, "Failed to bind gdb socket");
+        NGLOG_ERROR(Debug_GDBStub, "Failed to bind gdb socket");
    }
    if (listen(tmpsock, 1) < 0) {
-        LOG_ERROR(Debug_GDBStub, "Failed to listen to gdb socket");
+        NGLOG_ERROR(Debug_GDBStub, "Failed to listen to gdb socket");
    }
    // Wait for gdb to connect
-    LOG_INFO(Debug_GDBStub, "Waiting for gdb to connect...\n");
+    NGLOG_INFO(Debug_GDBStub, "Waiting for gdb to connect...");
    sockaddr_in saddr_client;
    sockaddr* client_addr = reinterpret_cast<sockaddr*>(&saddr_client);
    socklen_t client_addrlen = sizeof(saddr_client);
@@ -1027,9 +1026,9 @@ static void Init(u16 port) {
        halt_loop = false;
        step_loop = false;
-        LOG_ERROR(Debug_GDBStub, "Failed to accept gdb client");
+        NGLOG_ERROR(Debug_GDBStub, "Failed to accept gdb client");
    } else {
-        LOG_INFO(Debug_GDBStub, "Client connected.\n");
+        NGLOG_INFO(Debug_GDBStub, "Client connected.");
        saddr_client.sin_addr.s_addr = ntohl(saddr_client.sin_addr.s_addr);
    }
@@ -1048,7 +1047,7 @@ void Shutdown() {
        return;
    }
-    LOG_INFO(Debug_GDBStub, "Stopping GDB ...");
+    NGLOG_INFO(Debug_GDBStub, "Stopping GDB ...");
    if (gdbserver_socket != -1) {
        shutdown(gdbserver_socket, SHUT_RDWR);
        gdbserver_socket = -1;
@@ -1058,7 +1057,7 @@ void Shutdown() {
    WSACleanup();
 #endif
-    LOG_INFO(Debug_GDBStub, "GDB stopped.");
+    NGLOG_INFO(Debug_GDBStub, "GDB stopped.");
 }
 bool IsServerEnabled() {
diff --git a/src/core/hle/kernel/condition_variable.cpp b/src/core/hle/kernel/condition_variable.cpp
deleted file mode 100644
index a786d7f74..000000000
--- a/src/core/hle/kernel/condition_variable.cpp
+++ /dev/null
@@ -1,64 +0,0 @@
-// Copyright 2018 yuzu emulator team
-// Licensed under GPLv2 or any later version
-// Refer to the license.txt file included.
-#include "common/assert.h"
-#include "core/hle/kernel/condition_variable.h"
-#include "core/hle/kernel/errors.h"
-#include "core/hle/kernel/kernel.h"
-#include "core/hle/kernel/object_address_table.h"
-#include "core/hle/kernel/thread.h"
-namespace Kernel {
-ConditionVariable::ConditionVariable() {}
-ConditionVariable::~ConditionVariable() {}
-ResultVal<SharedPtr<ConditionVariable>> ConditionVariable::Create(VAddr guest_addr,
-                                                                  std::string name) {
-    SharedPtr<ConditionVariable> condition_variable(new ConditionVariable);
-    condition_variable->name = std::move(name);
-    condition_variable->guest_addr = guest_addr;
-    condition_variable->mutex_addr = 0;
-    // Condition variables are referenced by guest address, so track this in the kernel
-    g_object_address_table.Insert(guest_addr, condition_variable);
-    return MakeResult<SharedPtr<ConditionVariable>>(std::move(condition_variable));
-}
-bool ConditionVariable::ShouldWait(Thread* thread) const {
-    return GetAvailableCount() <= 0;
-}
-void ConditionVariable::Acquire(Thread* thread) {
-    if (GetAvailableCount() <= 0)
-        return;
-    SetAvailableCount(GetAvailableCount() - 1);
-}
-ResultCode ConditionVariable::Release(s32 target) {
-    if (target == -1) {
-        // When -1, wake up all waiting threads
-        SetAvailableCount(static_cast<s32>(GetWaitingThreads().size()));
-        WakeupAllWaitingThreads();
-    } else {
-        // Otherwise, wake up just a single thread
-        SetAvailableCount(target);
-        WakeupWaitingThread(GetHighestPriorityReadyThread());
-    }
-    return RESULT_SUCCESS;
-}
-s32 ConditionVariable::GetAvailableCount() const {
-    return Memory::Read32(guest_addr);
-}
-void ConditionVariable::SetAvailableCount(s32 value) const {
-    Memory::Write32(guest_addr, value);
-}
-} // namespace Kernel
diff --git a/src/core/hle/kernel/condition_variable.h b/src/core/hle/kernel/condition_variable.h
deleted file mode 100644
index 1c9f06769..000000000
--- a/src/core/hle/kernel/condition_variable.h
+++ /dev/null
@@ -1,63 +0,0 @@
-// Copyright 2018 yuzu emulator team
-// Licensed under GPLv2 or any later version
-// Refer to the license.txt file included.
-#pragma once
-#include <string>
-#include <queue>
-#include "common/common_types.h"
-#include "core/hle/kernel/kernel.h"
-#include "core/hle/kernel/wait_object.h"
-#include "core/hle/result.h"
-namespace Kernel {
-class ConditionVariable final : public WaitObject {
-public:
-    /**
-     * Creates a condition variable.
-     * @param guest_addr Address of the object tracking the condition variable in guest memory. If
-     * specified, this condition variable will update the guest object when its state changes.
-     * @param name Optional name of condition variable.
-     * @return The created condition variable.
-     */
-    static ResultVal<SharedPtr<ConditionVariable>> Create(VAddr guest_addr,
-                                                          std::string name = "Unknown");
-    std::string GetTypeName() const override {
-        return "ConditionVariable";
-    }
-    std::string GetName() const override {
-        return name;
-    }
-    static const HandleType HANDLE_TYPE = HandleType::ConditionVariable;
-    HandleType GetHandleType() const override {
-        return HANDLE_TYPE;
-    }
-    s32 GetAvailableCount() const;
-    void SetAvailableCount(s32 value) const;
-    std::string name; ///< Name of condition variable (optional)
-    VAddr guest_addr; ///< Address of the guest condition variable value
-    VAddr mutex_addr; ///< (optional) Address of guest mutex value associated with this condition
-                      ///< variable, used for implementing events
-    bool ShouldWait(Thread* thread) const override;
-    void Acquire(Thread* thread) override;
-    /**
-     * Releases a slot from a condition variable.
-     * @param target The number of threads to wakeup, -1 is all.
-     * @return ResultCode indicating if the operation succeeded.
-     */
-    ResultCode Release(s32 target);
-private:
-    ConditionVariable();
-    ~ConditionVariable() override;
-};
-} // namespace Kernel
diff --git a/src/core/hle/kernel/errors.h b/src/core/hle/kernel/errors.h
index 29d8dfdaa..5be20c878 100644
--- a/src/core/hle/kernel/errors.h
+++ b/src/core/hle/kernel/errors.h
@@ -20,6 +20,7 @@ enum {
    MaxConnectionsReached = 52,
    // Confirmed Switch OS error codes
+    MisalignedAddress = 102,
    InvalidHandle = 114,
    Timeout = 117,
    SynchronizationCanceled = 118,
diff --git a/src/core/hle/kernel/handle_table.cpp b/src/core/hle/kernel/handle_table.cpp
index 822449cd5..f7a9920d8 100644
--- a/src/core/hle/kernel/handle_table.cpp
+++ b/src/core/hle/kernel/handle_table.cpp
@@ -26,7 +26,7 @@ ResultVal<Handle> HandleTable::Create(SharedPtr<Object> obj) {
    u16 slot = next_free_slot;
    if (slot >= generations.size()) {
-        LOG_ERROR(Kernel, "Unable to allocate Handle, too many slots in use.");
+        NGLOG_ERROR(Kernel, "Unable to allocate Handle, too many slots in use.");
        return ERR_OUT_OF_HANDLES;
    }
    next_free_slot = generations[slot];
@@ -48,7 +48,7 @@ ResultVal<Handle> HandleTable::Create(SharedPtr<Object> obj) {
 ResultVal<Handle> HandleTable::Duplicate(Handle handle) {
    SharedPtr<Object> object = GetGeneric(handle);
    if (object == nullptr) {
-        LOG_ERROR(Kernel, "Tried to duplicate invalid handle: %08X", handle);
+        NGLOG_ERROR(Kernel, "Tried to duplicate invalid handle: {:08X}", handle);
        return ERR_INVALID_HANDLE;
    }
    return Create(std::move(object));
diff --git a/src/core/hle/kernel/hle_ipc.cpp b/src/core/hle/kernel/hle_ipc.cpp
index bef4f15f5..aa6ca1026 100644
--- a/src/core/hle/kernel/hle_ipc.cpp
+++ b/src/core/hle/kernel/hle_ipc.cpp
@@ -118,7 +118,7 @@ void HLERequestContext::ParseCommandBuffer(u32_le* src_cmdbuf, bool incoming) {
                std::make_shared<IPC::DomainMessageHeader>(rp.PopRaw<IPC::DomainMessageHeader>());
        } else {
            if (Session()->IsDomain())
-                LOG_WARNING(IPC, "Domain request has no DomainMessageHeader!");
+                NGLOG_WARNING(IPC, "Domain request has no DomainMessageHeader!");
        }
    }
@@ -270,7 +270,8 @@ size_t HLERequestContext::WriteBuffer(const void* buffer, size_t size) const {
    const bool is_buffer_b{BufferDescriptorB().size() && BufferDescriptorB()[0].Size()};
    const size_t buffer_size{GetWriteBufferSize()};
    if (size > buffer_size) {
-        LOG_CRITICAL(Core, "size (%016zx) is greater than buffer_size (%016zx)", size, buffer_size);
+        NGLOG_CRITICAL(Core, "size ({:016X}) is greater than buffer_size ({:016X})", size,
+                       buffer_size);
        size = buffer_size; // TODO(bunnei): This needs to be HW tested
    }
diff --git a/src/core/hle/kernel/kernel.h b/src/core/hle/kernel/kernel.h
index 053bf4e17..402ae900f 100644
--- a/src/core/hle/kernel/kernel.h
+++ b/src/core/hle/kernel/kernel.h
@@ -18,12 +18,10 @@ using Handle = u32;
 enum class HandleType : u32 {
    Unknown,
    Event,
-    Mutex,
    SharedMemory,
    Thread,
    Process,
    AddressArbiter,
-    ConditionVariable,
    Timer,
    ResourceLimit,
    CodeSet,
@@ -63,9 +61,7 @@ public:
    bool IsWaitable() const {
        switch (GetHandleType()) {
        case HandleType::Event:
-        case HandleType::Mutex:
        case HandleType::Thread:
-        case HandleType::ConditionVariable:
        case HandleType::Timer:
        case HandleType::ServerPort:
        case HandleType::ServerSession:
diff --git a/src/core/hle/kernel/mutex.cpp b/src/core/hle/kernel/mutex.cpp
index 0b9dc700c..63733ad79 100644
--- a/src/core/hle/kernel/mutex.cpp
+++ b/src/core/hle/kernel/mutex.cpp
@@ -7,6 +7,7 @@
 #include <boost/range/algorithm_ext/erase.hpp>
 #include "common/assert.h"
 #include "core/core.h"
+#include "core/hle/kernel/errors.h"
 #include "core/hle/kernel/handle_table.h"
 #include "core/hle/kernel/kernel.h"
 #include "core/hle/kernel/mutex.h"
@@ -15,124 +16,120 @@
 namespace Kernel {
-void ReleaseThreadMutexes(Thread* thread) {
+/// Returns the number of threads that are waiting for a mutex, and the highest priority one among
-    for (auto& mtx : thread->held_mutexes) {
+/// those.
-        mtx->SetHasWaiters(false);
+static std::pair<SharedPtr<Thread>, u32> GetHighestPriorityMutexWaitingThread(
-        mtx->SetHoldingThread(nullptr);
+    SharedPtr<Thread> current_thread, VAddr mutex_addr) {
-        mtx->WakeupAllWaitingThreads();
-    }
-    thread->held_mutexes.clear();
-}
-Mutex::Mutex() {}
+    SharedPtr<Thread> highest_priority_thread;
-Mutex::~Mutex() {}
+    u32 num_waiters = 0;
-SharedPtr<Mutex> Mutex::Create(SharedPtr<Kernel::Thread> holding_thread, VAddr guest_addr,
+    for (auto& thread : current_thread->wait_mutex_threads) {
-                               std::string name) {
+        if (thread->mutex_wait_address != mutex_addr)
-    SharedPtr<Mutex> mutex(new Mutex);
+            continue;
-    mutex->guest_addr = guest_addr;
+        ASSERT(thread->status == THREADSTATUS_WAIT_MUTEX);
-    mutex->name = std::move(name);
-    // If mutex was initialized with a holding thread, acquire it by the holding thread
+        ++num_waiters;
-    if (holding_thread) {
+        if (highest_priority_thread == nullptr ||
-        mutex->Acquire(holding_thread.get());
+            thread->GetPriority() < highest_priority_thread->GetPriority()) {
+            highest_priority_thread = thread;
+        }
    }
-    // Mutexes are referenced by guest address, so track this in the kernel
+    return {highest_priority_thread, num_waiters};
-    g_object_address_table.Insert(guest_addr, mutex);
-    return mutex;
 }
-bool Mutex::ShouldWait(Thread* thread) const {
+/// Update the mutex owner field of all threads waiting on the mutex to point to the new owner.
-    auto holding_thread = GetHoldingThread();
+static void TransferMutexOwnership(VAddr mutex_addr, SharedPtr<Thread> current_thread,
-    return holding_thread != nullptr && thread != holding_thread;
+                                   SharedPtr<Thread> new_owner) {
+    auto threads = current_thread->wait_mutex_threads;
+    for (auto& thread : threads) {
+        if (thread->mutex_wait_address != mutex_addr)
+            continue;
+        ASSERT(thread->lock_owner == current_thread);
+        current_thread->RemoveMutexWaiter(thread);
+        if (new_owner != thread)
+            new_owner->AddMutexWaiter(thread);
+    }
 }
-void Mutex::Acquire(Thread* thread) {
+ResultCode Mutex::TryAcquire(VAddr address, Handle holding_thread_handle,
-    ASSERT_MSG(!ShouldWait(thread), "object unavailable!");
+                             Handle requesting_thread_handle) {
+    // The mutex address must be 4-byte aligned
+    if ((address % sizeof(u32)) != 0) {
+        return ResultCode(ErrorModule::Kernel, ErrCodes::MisalignedAddress);
+    }
-    priority = thread->current_priority;
+    SharedPtr<Thread> holding_thread = g_handle_table.Get<Thread>(holding_thread_handle);
-    thread->held_mutexes.insert(this);
+    SharedPtr<Thread> requesting_thread = g_handle_table.Get<Thread>(requesting_thread_handle);
-    SetHoldingThread(thread);
-    thread->UpdatePriority();
-    Core::System::GetInstance().PrepareReschedule();
-}
-ResultCode Mutex::Release(Thread* thread) {
+    // TODO(Subv): It is currently unknown if it is possible to lock a mutex in behalf of another
-    auto holding_thread = GetHoldingThread();
+    // thread.
-    ASSERT(holding_thread);
+    ASSERT(requesting_thread == GetCurrentThread());
-    // We can only release the mutex if it's held by the calling thread.
+    u32 addr_value = Memory::Read32(address);
-    ASSERT(thread == holding_thread);
+    // If the mutex isn't being held, just return success.
+    if (addr_value != (holding_thread_handle | Mutex::MutexHasWaitersFlag)) {
+        return RESULT_SUCCESS;
+    }
+    if (holding_thread == nullptr)
+        return ERR_INVALID_HANDLE;
+    // Wait until the mutex is released
+    GetCurrentThread()->mutex_wait_address = address;
+    GetCurrentThread()->wait_handle = requesting_thread_handle;
+    GetCurrentThread()->status = THREADSTATUS_WAIT_MUTEX;
+    GetCurrentThread()->wakeup_callback = nullptr;
+    // Update the lock holder thread's priority to prevent priority inversion.
+    holding_thread->AddMutexWaiter(GetCurrentThread());
-    holding_thread->held_mutexes.erase(this);
-    holding_thread->UpdatePriority();
-    SetHoldingThread(nullptr);
-    SetHasWaiters(!GetWaitingThreads().empty());
-    WakeupAllWaitingThreads();
    Core::System::GetInstance().PrepareReschedule();
    return RESULT_SUCCESS;
 }
-void Mutex::AddWaitingThread(SharedPtr<Thread> thread) {
+ResultCode Mutex::Release(VAddr address) {
-    WaitObject::AddWaitingThread(thread);
+    // The mutex address must be 4-byte aligned
-    thread->pending_mutexes.insert(this);
+    if ((address % sizeof(u32)) != 0) {
-    SetHasWaiters(true);
+        return ResultCode(ErrorModule::Kernel, ErrCodes::MisalignedAddress);
-    UpdatePriority();
+    }
-}
-void Mutex::RemoveWaitingThread(Thread* thread) {
-    WaitObject::RemoveWaitingThread(thread);
-    thread->pending_mutexes.erase(this);
-    if (!GetHasWaiters())
-        SetHasWaiters(!GetWaitingThreads().empty());
-    UpdatePriority();
-}
-void Mutex::UpdatePriority() {
+    auto [thread, num_waiters] = GetHighestPriorityMutexWaitingThread(GetCurrentThread(), address);
-    if (!GetHoldingThread())
-        return;
-    u32 best_priority = THREADPRIO_LOWEST;
+    // There are no more threads waiting for the mutex, release it completely.
-    for (auto& waiter : GetWaitingThreads()) {
+    if (thread == nullptr) {
-        if (waiter->current_priority < best_priority)
+        ASSERT(GetCurrentThread()->wait_mutex_threads.empty());
-            best_priority = waiter->current_priority;
+        Memory::Write32(address, 0);
+        return RESULT_SUCCESS;
    }
-    if (best_priority != priority) {
+    // Transfer the ownership of the mutex from the previous owner to the new one.
-        priority = best_priority;
+    TransferMutexOwnership(address, GetCurrentThread(), thread);
-        GetHoldingThread()->UpdatePriority();
-    }
-}
-Handle Mutex::GetOwnerHandle() const {
+    u32 mutex_value = thread->wait_handle;
-    GuestState guest_state{Memory::Read32(guest_addr)};
-    return guest_state.holding_thread_handle;
-}
-SharedPtr<Thread> Mutex::GetHoldingThread() const {
+    if (num_waiters >= 2) {
-    GuestState guest_state{Memory::Read32(guest_addr)};
+        // Notify the guest that there are still some threads waiting for the mutex
-    return g_handle_table.Get<Thread>(guest_state.holding_thread_handle);
+        mutex_value |= Mutex::MutexHasWaitersFlag;
-}
+    }
-void Mutex::SetHoldingThread(SharedPtr<Thread> thread) {
+    // Grant the mutex to the next waiting thread and resume it.
-    GuestState guest_state{Memory::Read32(guest_addr)};
+    Memory::Write32(address, mutex_value);
-    guest_state.holding_thread_handle.Assign(thread ? thread->guest_handle : 0);
-    Memory::Write32(guest_addr, guest_state.raw);
-}
-bool Mutex::GetHasWaiters() const {
+    ASSERT(thread->status == THREADSTATUS_WAIT_MUTEX);
-    GuestState guest_state{Memory::Read32(guest_addr)};
+    thread->ResumeFromWait();
-    return guest_state.has_waiters != 0;
-}
-void Mutex::SetHasWaiters(bool has_waiters) {
+    thread->lock_owner = nullptr;
-    GuestState guest_state{Memory::Read32(guest_addr)};
+    thread->condvar_wait_address = 0;
-    guest_state.has_waiters.Assign(has_waiters ? 1 : 0);
+    thread->mutex_wait_address = 0;
-    Memory::Write32(guest_addr, guest_state.raw);
+    thread->wait_handle = 0;
-}
+    return RESULT_SUCCESS;
+}
 } // namespace Kernel
diff --git a/src/core/hle/kernel/mutex.h b/src/core/hle/kernel/mutex.h
index 38db21005..3117e7c70 100644
--- a/src/core/hle/kernel/mutex.h
+++ b/src/core/hle/kernel/mutex.h
@@ -15,87 +15,23 @@ namespace Kernel {
 class Thread;
-class Mutex final : public WaitObject {
+class Mutex final {
 public:
-    /**
+    /// Flag that indicates that a mutex still has threads waiting for it.
-     * Creates a mutex.
+    static constexpr u32 MutexHasWaitersFlag = 0x40000000;
-     * @param holding_thread Specifies a thread already holding the mutex. If not nullptr, this
+    /// Mask of the bits in a mutex address value that contain the mutex owner.
-     * thread will acquire the mutex.
+    static constexpr u32 MutexOwnerMask = 0xBFFFFFFF;
-     * @param guest_addr Address of the object tracking the mutex in guest memory. If specified,
-     * this mutex will update the guest object when its state changes.
-     * @param name Optional name of mutex
-     * @return Pointer to new Mutex object
-     */
-    static SharedPtr<Mutex> Create(SharedPtr<Kernel::Thread> holding_thread, VAddr guest_addr = 0,
-                                   std::string name = "Unknown");
-    std::string GetTypeName() const override {
+    /// Attempts to acquire a mutex at the specified address.
-        return "Mutex";
+    static ResultCode TryAcquire(VAddr address, Handle holding_thread_handle,
-    }
+                                 Handle requesting_thread_handle);
-    std::string GetName() const override {
-        return name;
-    }
-    static const HandleType HANDLE_TYPE = HandleType::Mutex;
+    /// Releases the mutex at the specified address.
-    HandleType GetHandleType() const override {
+    static ResultCode Release(VAddr address);
-        return HANDLE_TYPE;
-    }
-    u32 priority;     ///< The priority of the mutex, used for priority inheritance.
-    std::string name; ///< Name of mutex (optional)
-    VAddr guest_addr; ///< Address of the guest mutex value
-    /**
-     * Elevate the mutex priority to the best priority
-     * among the priorities of all its waiting threads.
-     */
-    void UpdatePriority();
-    bool ShouldWait(Thread* thread) const override;
-    void Acquire(Thread* thread) override;
-    void AddWaitingThread(SharedPtr<Thread> thread) override;
-    void RemoveWaitingThread(Thread* thread) override;
-    /**
-     * Attempts to release the mutex from the specified thread.
-     * @param thread Thread that wants to release the mutex.
-     * @returns The result code of the operation.
-     */
-    ResultCode Release(Thread* thread);
-    /// Gets the handle to the holding process stored in the guest state.
-    Handle GetOwnerHandle() const;
-    /// Gets the Thread pointed to by the owner handle
-    SharedPtr<Thread> GetHoldingThread() const;
-    /// Sets the holding process handle in the guest state.
-    void SetHoldingThread(SharedPtr<Thread> thread);
-    /// Returns the has_waiters bit in the guest state.
-    bool GetHasWaiters() const;
-    /// Sets the has_waiters bit in the guest state.
-    void SetHasWaiters(bool has_waiters);
 private:
-    Mutex();
+    Mutex() = default;
-    ~Mutex() override;
+    ~Mutex() = default;
-    /// Object in guest memory used to track the mutex state
-    union GuestState {
-        u32_le raw;
-        /// Handle of the thread that currently holds the mutex, 0 if available
-        BitField<0, 30, u32_le> holding_thread_handle;
-        /// 1 when there are threads waiting for this mutex, otherwise 0
-        BitField<30, 1, u32_le> has_waiters;
-    };
-    static_assert(sizeof(GuestState) == 4, "GuestState size is incorrect");
 };
-/**
- * Releases all the mutexes held by the specified thread
- * @param thread Thread that is holding the mutexes
- */
-void ReleaseThreadMutexes(Thread* thread);
 } // namespace Kernel
diff --git a/src/core/hle/kernel/process.cpp b/src/core/hle/kernel/process.cpp
index 2cffec198..751a0524d 100644
--- a/src/core/hle/kernel/process.cpp
+++ b/src/core/hle/kernel/process.cpp
@@ -54,7 +54,7 @@ void Process::ParseKernelCaps(const u32* kernel_caps, size_t len) {
            continue;
        } else if ((type & 0xF00) == 0xE00) { // 0x0FFF
            // Allowed interrupts list
-            LOG_WARNING(Loader, "ExHeader allowed interrupts list ignored");
+            NGLOG_WARNING(Loader, "ExHeader allowed interrupts list ignored");
        } else if ((type & 0xF80) == 0xF00) { // 0x07FF
            // Allowed syscalls mask
            unsigned int index = ((descriptor >> 24) & 7) * 24;
@@ -74,7 +74,7 @@ void Process::ParseKernelCaps(const u32* kernel_caps, size_t len) {
        } else if ((type & 0xFFE) == 0xFF8) { // 0x001F
            // Mapped memory range
            if (i + 1 >= len || ((kernel_caps[i + 1] >> 20) & 0xFFE) != 0xFF8) {
-                LOG_WARNING(Loader, "Incomplete exheader memory range descriptor ignored.");
+                NGLOG_WARNING(Loader, "Incomplete exheader memory range descriptor ignored.");
                continue;
            }
            u32 end_desc = kernel_caps[i + 1];
@@ -109,9 +109,9 @@ void Process::ParseKernelCaps(const u32* kernel_caps, size_t len) {
            int minor = kernel_version & 0xFF;
            int major = (kernel_version >> 8) & 0xFF;
-            LOG_INFO(Loader, "ExHeader kernel version: %d.%d", major, minor);
+            NGLOG_INFO(Loader, "ExHeader kernel version: {}.{}", major, minor);
        } else {
-            LOG_ERROR(Loader, "Unhandled kernel caps descriptor: 0x%08X", descriptor);
+            NGLOG_ERROR(Loader, "Unhandled kernel caps descriptor: {:#010X}", descriptor);
        }
    }
 }
diff --git a/src/core/hle/kernel/resource_limit.cpp b/src/core/hle/kernel/resource_limit.cpp
index 88ca8ad7e..0ef5fc57d 100644
--- a/src/core/hle/kernel/resource_limit.cpp
+++ b/src/core/hle/kernel/resource_limit.cpp
@@ -29,7 +29,7 @@ SharedPtr<ResourceLimit> ResourceLimit::GetForCategory(ResourceLimitCategory cat
    case ResourceLimitCategory::OTHER:
        return resource_limits[static_cast<u8>(category)];
    default:
-        LOG_CRITICAL(Kernel, "Unknown resource limit category");
+        NGLOG_CRITICAL(Kernel, "Unknown resource limit category");
        UNREACHABLE();
    }
 }
@@ -55,7 +55,7 @@ s32 ResourceLimit::GetCurrentResourceValue(ResourceType resource) const {
    case ResourceType::CPUTime:
        return current_cpu_time;
    default:
-        LOG_ERROR(Kernel, "Unknown resource type=%08X", static_cast<u32>(resource));
+        NGLOG_ERROR(Kernel, "Unknown resource type={:08X}", static_cast<u32>(resource));
        UNIMPLEMENTED();
        return 0;
    }
@@ -84,7 +84,7 @@ u32 ResourceLimit::GetMaxResourceValue(ResourceType resource) const {
    case ResourceType::CPUTime:
        return max_cpu_time;
    default:
-        LOG_ERROR(Kernel, "Unknown resource type=%08X", static_cast<u32>(resource));
+        NGLOG_ERROR(Kernel, "Unknown resource type={:08X}", static_cast<u32>(resource));
        UNIMPLEMENTED();
        return 0;
    }
diff --git a/src/core/hle/kernel/scheduler.cpp b/src/core/hle/kernel/scheduler.cpp
index 921f27efb..ff6a0941a 100644
--- a/src/core/hle/kernel/scheduler.cpp
+++ b/src/core/hle/kernel/scheduler.cpp
@@ -94,11 +94,11 @@ void Scheduler::Reschedule() {
    Thread* next = PopNextReadyThread();
    if (cur && next) {
-        LOG_TRACE(Kernel, "context switch %u -> %u", cur->GetObjectId(), next->GetObjectId());
+        NGLOG_TRACE(Kernel, "context switch {} -> {}", cur->GetObjectId(), next->GetObjectId());
    } else if (cur) {
-        LOG_TRACE(Kernel, "context switch %u -> idle", cur->GetObjectId());
+        NGLOG_TRACE(Kernel, "context switch {} -> idle", cur->GetObjectId());
    } else if (next) {
-        LOG_TRACE(Kernel, "context switch idle -> %u", next->GetObjectId());
+        NGLOG_TRACE(Kernel, "context switch idle -> {}", next->GetObjectId());
    }
    SwitchContext(next);
diff --git a/src/core/hle/kernel/server_session.cpp b/src/core/hle/kernel/server_session.cpp
index 33397d84f..b1f8e771c 100644
--- a/src/core/hle/kernel/server_session.cpp
+++ b/src/core/hle/kernel/server_session.cpp
@@ -68,7 +68,7 @@ ResultCode ServerSession::HandleDomainSyncRequest(Kernel::HLERequestContext& con
            return domain_request_handlers[object_id - 1]->HandleSyncRequest(context);
        case IPC::DomainMessageHeader::CommandType::CloseVirtualHandle: {
-            LOG_DEBUG(IPC, "CloseVirtualHandle, object_id=0x%08X", object_id);
+            NGLOG_DEBUG(IPC, "CloseVirtualHandle, object_id={:#010X}", object_id);
            domain_request_handlers[object_id - 1] = nullptr;
@@ -78,8 +78,8 @@ ResultCode ServerSession::HandleDomainSyncRequest(Kernel::HLERequestContext& con
        }
        }
-        LOG_CRITICAL(IPC, "Unknown domain command=%d",
+        NGLOG_CRITICAL(IPC, "Unknown domain command={}",
-                     static_cast<int>(domain_message_header->command.Value()));
+                       static_cast<int>(domain_message_header->command.Value()));
        ASSERT(false);
    }
diff --git a/src/core/hle/kernel/shared_memory.cpp b/src/core/hle/kernel/shared_memory.cpp
index bc99993c8..f0b65c73d 100644
--- a/src/core/hle/kernel/shared_memory.cpp
+++ b/src/core/hle/kernel/shared_memory.cpp
@@ -107,16 +107,16 @@ ResultCode SharedMemory::Map(Process* target_process, VAddr address, MemoryPermi
    // Error out if the requested permissions don't match what the creator process allows.
    if (static_cast<u32>(permissions) & ~static_cast<u32>(own_other_permissions)) {
-        LOG_ERROR(Kernel, "cannot map id=%u, address=0x%lx name=%s, permissions don't match",
+        NGLOG_ERROR(Kernel, "cannot map id={}, address={:#X} name={}, permissions don't match",
-                  GetObjectId(), address, name.c_str());
+                    GetObjectId(), address, name);
        return ERR_INVALID_COMBINATION;
    }
    // Error out if the provided permissions are not compatible with what the creator process needs.
    if (other_permissions != MemoryPermission::DontCare &&
        static_cast<u32>(this->permissions) & ~static_cast<u32>(other_permissions)) {
-        LOG_ERROR(Kernel, "cannot map id=%u, address=0x%lx name=%s, permissions don't match",
+        NGLOG_ERROR(Kernel, "cannot map id={}, address={:#X} name={}, permissions don't match",
-                  GetObjectId(), address, name.c_str());
+                    GetObjectId(), address, name);
        return ERR_WRONG_PERMISSION;
    }
@@ -131,9 +131,10 @@ ResultCode SharedMemory::Map(Process* target_process, VAddr address, MemoryPermi
    auto result = target_process->vm_manager.MapMemoryBlock(
        target_address, backing_block, backing_block_offset, size, MemoryState::Shared);
    if (result.Failed()) {
-        LOG_ERROR(Kernel,
+        NGLOG_ERROR(
-                  "cannot map id=%u, target_address=0x%lx name=%s, error mapping to virtual memory",
+            Kernel,
-                  GetObjectId(), target_address, name.c_str());
+            "cannot map id={}, target_address={:#X} name={}, error mapping to virtual memory",
+            GetObjectId(), target_address, name);
        return result.Code();
    }
@@ -151,7 +152,7 @@ VMAPermission SharedMemory::ConvertPermissions(MemoryPermission permission) {
    u32 masked_permissions =
        static_cast<u32>(permission) & static_cast<u32>(MemoryPermission::ReadWriteExecute);
    return static_cast<VMAPermission>(masked_permissions);
-};
+}
 u8* SharedMemory::GetPointer(u32 offset) {
    return backing_block->data() + backing_block_offset + offset;
diff --git a/src/core/hle/kernel/svc.cpp b/src/core/hle/kernel/svc.cpp
index 633740992..cb19b1a69 100644
--- a/src/core/hle/kernel/svc.cpp
+++ b/src/core/hle/kernel/svc.cpp
@@ -13,7 +13,6 @@
 #include "core/core_timing.h"
 #include "core/hle/kernel/client_port.h"
 #include "core/hle/kernel/client_session.h"
-#include "core/hle/kernel/condition_variable.h"
 #include "core/hle/kernel/event.h"
 #include "core/hle/kernel/handle_table.h"
 #include "core/hle/kernel/mutex.h"
@@ -32,7 +31,7 @@ namespace Kernel {
 /// Set the process heap to a given Size. It can both extend and shrink the heap.
 static ResultCode SetHeapSize(VAddr* heap_addr, u64 heap_size) {
-    LOG_TRACE(Kernel_SVC, "called, heap_size=0x%llx", heap_size);
+    NGLOG_TRACE(Kernel_SVC, "called, heap_size={:#X}", heap_size);
    auto& process = *Core::CurrentProcess();
    CASCADE_RESULT(*heap_addr,
                   process.HeapAllocate(Memory::HEAP_VADDR, heap_size, VMAPermission::ReadWrite));
@@ -40,21 +39,21 @@ static ResultCode SetHeapSize(VAddr* heap_addr, u64 heap_size) {
 }
 static ResultCode SetMemoryAttribute(VAddr addr, u64 size, u32 state0, u32 state1) {
-    LOG_WARNING(Kernel_SVC, "(STUBBED) called, addr=0x%lx", addr);
+    NGLOG_WARNING(Kernel_SVC, "(STUBBED) called, addr={:#X}", addr);
    return RESULT_SUCCESS;
 }
 /// Maps a memory range into a different range.
 static ResultCode MapMemory(VAddr dst_addr, VAddr src_addr, u64 size) {
-    LOG_TRACE(Kernel_SVC, "called, dst_addr=0x%llx, src_addr=0x%llx, size=0x%llx", dst_addr,
+    NGLOG_TRACE(Kernel_SVC, "called, dst_addr={:#X}, src_addr={:#X}, size={:#X}", dst_addr,
-              src_addr, size);
+                src_addr, size);
    return Core::CurrentProcess()->MirrorMemory(dst_addr, src_addr, size);
 }
 /// Unmaps a region that was previously mapped with svcMapMemory
 static ResultCode UnmapMemory(VAddr dst_addr, VAddr src_addr, u64 size) {
-    LOG_TRACE(Kernel_SVC, "called, dst_addr=0x%llx, src_addr=0x%llx, size=0x%llx", dst_addr,
+    NGLOG_TRACE(Kernel_SVC, "called, dst_addr={:#X}, src_addr={:#X}, size={:#X}", dst_addr,
-              src_addr, size);
+                src_addr, size);
    return Core::CurrentProcess()->UnmapMemory(dst_addr, src_addr, size);
 }
@@ -69,11 +68,11 @@ static ResultCode ConnectToNamedPort(Handle* out_handle, VAddr port_name_address
    if (port_name.size() > PortNameMaxLength)
        return ERR_PORT_NAME_TOO_LONG;
-    LOG_TRACE(Kernel_SVC, "called port_name=%s", port_name.c_str());
+    NGLOG_TRACE(Kernel_SVC, "called port_name={}", port_name);
    auto it = Service::g_kernel_named_ports.find(port_name);
    if (it == Service::g_kernel_named_ports.end()) {
-        LOG_WARNING(Kernel_SVC, "tried to connect to unknown port: %s", port_name.c_str());
+        NGLOG_WARNING(Kernel_SVC, "tried to connect to unknown port: {}", port_name);
        return ERR_NOT_FOUND;
    }
@@ -91,11 +90,11 @@ static ResultCode ConnectToNamedPort(Handle* out_handle, VAddr port_name_address
 static ResultCode SendSyncRequest(Handle handle) {
    SharedPtr<ClientSession> session = g_handle_table.Get<ClientSession>(handle);
    if (!session) {
-        LOG_ERROR(Kernel_SVC, "called with invalid handle=0x%08X", handle);
+        NGLOG_ERROR(Kernel_SVC, "called with invalid handle={:#010X}", handle);
        return ERR_INVALID_HANDLE;
    }
-    LOG_TRACE(Kernel_SVC, "called handle=0x%08X(%s)", handle, session->GetName().c_str());
+    NGLOG_TRACE(Kernel_SVC, "called handle={:#010X}({})", handle, session->GetName());
    Core::System::GetInstance().PrepareReschedule();
@@ -106,7 +105,7 @@ static ResultCode SendSyncRequest(Handle handle) {
 /// Get the ID for the specified thread.
 static ResultCode GetThreadId(u32* thread_id, Handle thread_handle) {
-    LOG_TRACE(Kernel_SVC, "called thread=0x%08X", thread_handle);
+    NGLOG_TRACE(Kernel_SVC, "called thread={:#010X}", thread_handle);
    const SharedPtr<Thread> thread = g_handle_table.Get<Thread>(thread_handle);
    if (!thread) {
@@ -119,7 +118,7 @@ static ResultCode GetThreadId(u32* thread_id, Handle thread_handle) {
 /// Get the ID of the specified process
 static ResultCode GetProcessId(u32* process_id, Handle process_handle) {
-    LOG_TRACE(Kernel_SVC, "called process=0x%08X", process_handle);
+    NGLOG_TRACE(Kernel_SVC, "called process={:#010X}", process_handle);
    const SharedPtr<Process> process = g_handle_table.Get<Process>(process_handle);
    if (!process) {
@@ -179,8 +178,8 @@ static ResultCode WaitSynchronization1(
 /// Wait for the given handles to synchronize, timeout after the specified nanoseconds
 static ResultCode WaitSynchronization(Handle* index, VAddr handles_address, u64 handle_count,
                                      s64 nano_seconds) {
-    LOG_TRACE(Kernel_SVC, "called handles_address=0x%llx, handle_count=%d, nano_seconds=%d",
+    NGLOG_TRACE(Kernel_SVC, "called handles_address={:#X}, handle_count={}, nano_seconds={}",
-              handles_address, handle_count, nano_seconds);
+                handles_address, handle_count, nano_seconds);
    if (!Memory::IsValidVirtualAddress(handles_address))
        return ERR_INVALID_POINTER;
@@ -240,7 +239,7 @@ static ResultCode WaitSynchronization(Handle* index, VAddr handles_address, u64
 /// Resumes a thread waiting on WaitSynchronization
 static ResultCode CancelSynchronization(Handle thread_handle) {
-    LOG_TRACE(Kernel_SVC, "called thread=0x%08X", thread_handle);
+    NGLOG_TRACE(Kernel_SVC, "called thread={:#X}", thread_handle);
    const SharedPtr<Thread> thread = g_handle_table.Get<Thread>(thread_handle);
    if (!thread) {
@@ -257,56 +256,38 @@ static ResultCode CancelSynchronization(Handle thread_handle) {
 /// Attempts to locks a mutex, creating it if it does not already exist
 static ResultCode ArbitrateLock(Handle holding_thread_handle, VAddr mutex_addr,
                                Handle requesting_thread_handle) {
-    LOG_TRACE(Kernel_SVC,
+    NGLOG_TRACE(Kernel_SVC,
-              "called holding_thread_handle=0x%08X, mutex_addr=0x%llx, "
+                "called holding_thread_handle={:#010X}, mutex_addr={:#X}, "
-              "requesting_current_thread_handle=0x%08X",
+                "requesting_current_thread_handle={:#010X}",
-              holding_thread_handle, mutex_addr, requesting_thread_handle);
+                holding_thread_handle, mutex_addr, requesting_thread_handle);
-    SharedPtr<Thread> holding_thread = g_handle_table.Get<Thread>(holding_thread_handle);
-    SharedPtr<Thread> requesting_thread = g_handle_table.Get<Thread>(requesting_thread_handle);
-    ASSERT(requesting_thread);
-    ASSERT(requesting_thread == GetCurrentThread());
-    SharedPtr<Mutex> mutex = g_object_address_table.Get<Mutex>(mutex_addr);
-    if (!mutex) {
-        // Create a new mutex for the specified address if one does not already exist
-        mutex = Mutex::Create(holding_thread, mutex_addr);
-        mutex->name = Common::StringFromFormat("mutex-%llx", mutex_addr);
-    }
-    ASSERT(holding_thread == mutex->GetHoldingThread());
-    return WaitSynchronization1(mutex, requesting_thread.get());
+    return Mutex::TryAcquire(mutex_addr, holding_thread_handle, requesting_thread_handle);
 }
 /// Unlock a mutex
 static ResultCode ArbitrateUnlock(VAddr mutex_addr) {
-    LOG_TRACE(Kernel_SVC, "called mutex_addr=0x%llx", mutex_addr);
+    NGLOG_TRACE(Kernel_SVC, "called mutex_addr={:#X}", mutex_addr);
-    SharedPtr<Mutex> mutex = g_object_address_table.Get<Mutex>(mutex_addr);
+    return Mutex::Release(mutex_addr);
-    ASSERT(mutex);
-    return mutex->Release(GetCurrentThread());
 }
 /// Break program execution
 static void Break(u64 unk_0, u64 unk_1, u64 unk_2) {
-    LOG_CRITICAL(Debug_Emulated, "Emulated program broke execution!");
+    NGLOG_CRITICAL(Debug_Emulated, "Emulated program broke execution!");
    ASSERT(false);
 }
 /// Used to output a message on a debug hardware unit - does nothing on a retail unit
 static void OutputDebugString(VAddr address, s32 len) {
-    std::vector<char> string(len);
+    std::string str(len, '\0');
-    Memory::ReadBlock(address, string.data(), len);
+    Memory::ReadBlock(address, str.data(), str.size());
-    LOG_DEBUG(Debug_Emulated, "%.*s", len, string.data());
+    NGLOG_DEBUG(Debug_Emulated, "{}", str);
 }
 /// Gets system/memory information for the current process
 static ResultCode GetInfo(u64* result, u64 info_id, u64 handle, u64 info_sub_id) {
-    LOG_TRACE(Kernel_SVC, "called info_id=0x%X, info_sub_id=0x%X, handle=0x%08X", info_id,
+    NGLOG_TRACE(Kernel_SVC, "called info_id={:#X}, info_sub_id={:#X}, handle={:#010X}", info_id,
-              info_sub_id, handle);
+                info_sub_id, handle);
    auto& vm_manager = Core::CurrentProcess()->vm_manager;
@@ -357,12 +338,12 @@ static ResultCode GetInfo(u64* result, u64 info_id, u64 handle, u64 info_sub_id)
        *result = Core::CurrentProcess()->is_virtual_address_memory_enabled;
        break;
    case GetInfoType::TitleId:
-        LOG_WARNING(Kernel_SVC, "(STUBBED) Attempted to query titleid, returned 0");
+        NGLOG_WARNING(Kernel_SVC, "(STUBBED) Attempted to query titleid, returned 0");
        *result = 0;
        break;
    case GetInfoType::PrivilegedProcessId:
-        LOG_WARNING(Kernel_SVC,
+        NGLOG_WARNING(Kernel_SVC,
-                    "(STUBBED) Attempted to query priviledged process id bounds, returned 0");
+                      "(STUBBED) Attempted to query privileged process id bounds, returned 0");
        *result = 0;
        break;
    default:
@@ -374,13 +355,14 @@ static ResultCode GetInfo(u64* result, u64 info_id, u64 handle, u64 info_sub_id)
 /// Sets the thread activity
 static ResultCode SetThreadActivity(Handle handle, u32 unknown) {
-    LOG_WARNING(Kernel_SVC, "(STUBBED) called, handle=0x%08X, unknown=0x%08X", handle, unknown);
+    NGLOG_WARNING(Kernel_SVC, "(STUBBED) called, handle={:#010X}, unknown={:#010X}", handle,
+                  unknown);
    return RESULT_SUCCESS;
 }
 /// Gets the thread context
 static ResultCode GetThreadContext(Handle handle, VAddr addr) {
-    LOG_WARNING(Kernel_SVC, "(STUBBED) called, handle=0x%08X, addr=0x%" PRIx64, handle, addr);
+    NGLOG_WARNING(Kernel_SVC, "(STUBBED) called, handle={:#010X}, addr={:#X}", handle, addr);
    return RESULT_SUCCESS;
 }
@@ -412,11 +394,6 @@ static ResultCode SetThreadPriority(Handle handle, u32 priority) {
    }
    thread->SetPriority(priority);
-    thread->UpdatePriority();
-    // Update the mutexes that this thread is waiting for
-    for (auto& mutex : thread->pending_mutexes)
-        mutex->UpdatePriority();
    Core::System::GetInstance().PrepareReschedule();
    return RESULT_SUCCESS;
@@ -424,15 +401,15 @@ static ResultCode SetThreadPriority(Handle handle, u32 priority) {
 /// Get which CPU core is executing the current thread
 static u32 GetCurrentProcessorNumber() {
-    LOG_WARNING(Kernel_SVC, "(STUBBED) called, defaulting to processor 0");
+    NGLOG_WARNING(Kernel_SVC, "(STUBBED) called, defaulting to processor 0");
    return 0;
 }
 static ResultCode MapSharedMemory(Handle shared_memory_handle, VAddr addr, u64 size,
                                  u32 permissions) {
-    LOG_TRACE(Kernel_SVC,
+    NGLOG_TRACE(Kernel_SVC,
-              "called, shared_memory_handle=0x%08X, addr=0x%llx, size=0x%llx, permissions=0x%08X",
+                "called, shared_memory_handle={:#X}, addr={:#X}, size={:#X}, permissions={:#010X}",
-              shared_memory_handle, addr, size, permissions);
+                shared_memory_handle, addr, size, permissions);
    SharedPtr<SharedMemory> shared_memory = g_handle_table.Get<SharedMemory>(shared_memory_handle);
    if (!shared_memory) {
@@ -452,16 +429,15 @@ static ResultCode MapSharedMemory(Handle shared_memory_handle, VAddr addr, u64 s
        return shared_memory->Map(Core::CurrentProcess().get(), addr, permissions_type,
                                  MemoryPermission::DontCare);
    default:
-        LOG_ERROR(Kernel_SVC, "unknown permissions=0x%08X", permissions);
+        NGLOG_ERROR(Kernel_SVC, "unknown permissions={:#010X}", permissions);
    }
    return RESULT_SUCCESS;
 }
 static ResultCode UnmapSharedMemory(Handle shared_memory_handle, VAddr addr, u64 size) {
-    LOG_WARNING(Kernel_SVC,
+    NGLOG_WARNING(Kernel_SVC, "called, shared_memory_handle={:#010X}, addr={:#X}, size={:#X}",
-                "called, shared_memory_handle=0x%08X, addr=0x%" PRIx64 ", size=0x%" PRIx64 "",
+                  shared_memory_handle, addr, size);
-                shared_memory_handle, addr, size);
    SharedPtr<SharedMemory> shared_memory = g_handle_table.Get<SharedMemory>(shared_memory_handle);
@@ -489,19 +465,19 @@ static ResultCode QueryProcessMemory(MemoryInfo* memory_info, PageInfo* /*page_i
        memory_info->type = static_cast<u32>(vma->second.meminfo_state);
    }
-    LOG_TRACE(Kernel_SVC, "called process=0x%08X addr=%llx", process_handle, addr);
+    NGLOG_TRACE(Kernel_SVC, "called process={:#010X} addr={:X}", process_handle, addr);
    return RESULT_SUCCESS;
 }
 /// Query memory
 static ResultCode QueryMemory(MemoryInfo* memory_info, PageInfo* page_info, VAddr addr) {
-    LOG_TRACE(Kernel_SVC, "called, addr=%llx", addr);
+    NGLOG_TRACE(Kernel_SVC, "called, addr={:X}", addr);
    return QueryProcessMemory(memory_info, page_info, CurrentProcess, addr);
 }
 /// Exits the current process
 static void ExitProcess() {
-    LOG_INFO(Kernel_SVC, "Process %u exiting", Core::CurrentProcess()->process_id);
+    NGLOG_INFO(Kernel_SVC, "Process {} exiting", Core::CurrentProcess()->process_id);
    ASSERT_MSG(Core::CurrentProcess()->status == ProcessStatus::Running,
               "Process has already exited");
@@ -558,9 +534,9 @@ static ResultCode CreateThread(Handle* out_handle, VAddr entry_point, u64 arg, V
    case THREADPROCESSORID_2:
    case THREADPROCESSORID_3:
        // TODO(bunnei): Implement support for other processor IDs
-        LOG_ERROR(Kernel_SVC,
+        NGLOG_ERROR(Kernel_SVC,
-                  "Newly created thread must run in another thread (%u), unimplemented.",
+                    "Newly created thread must run in another thread ({}), unimplemented.",
-                  processor_id);
+                    processor_id);
        break;
    default:
        ASSERT_MSG(false, "Unsupported thread processor ID: %d", processor_id);
@@ -575,17 +551,17 @@ static ResultCode CreateThread(Handle* out_handle, VAddr entry_point, u64 arg, V
    Core::System::GetInstance().PrepareReschedule();
-    LOG_TRACE(Kernel_SVC,
+    NGLOG_TRACE(Kernel_SVC,
-              "called entrypoint=0x%08X (%s), arg=0x%08X, stacktop=0x%08X, "
+                "called entrypoint={:#010X} ({}), arg={:#010X}, stacktop={:#010X}, "
-              "threadpriority=0x%08X, processorid=0x%08X : created handle=0x%08X",
+                "threadpriority={:#010X}, processorid={:#010X} : created handle={:#010X}",
-              entry_point, name.c_str(), arg, stack_top, priority, processor_id, *out_handle);
+                entry_point, name, arg, stack_top, priority, processor_id, *out_handle);
    return RESULT_SUCCESS;
 }
 /// Starts the thread for the provided handle
 static ResultCode StartThread(Handle thread_handle) {
-    LOG_TRACE(Kernel_SVC, "called thread=0x%08X", thread_handle);
+    NGLOG_TRACE(Kernel_SVC, "called thread={:#010X}", thread_handle);
    const SharedPtr<Thread> thread = g_handle_table.Get<Thread>(thread_handle);
    if (!thread) {
@@ -599,7 +575,7 @@ static ResultCode StartThread(Handle thread_handle) {
 /// Called when a thread exits
 static void ExitThread() {
-    LOG_TRACE(Kernel_SVC, "called, pc=0x%08X", Core::CPU().GetPC());
+    NGLOG_TRACE(Kernel_SVC, "called, pc={:#010X}", Core::CPU().GetPC());
    ExitCurrentThread();
    Core::System::GetInstance().PrepareReschedule();
@@ -607,7 +583,7 @@ static void ExitThread() {
 /// Sleep the current thread
 static void SleepThread(s64 nanoseconds) {
-    LOG_TRACE(Kernel_SVC, "called nanoseconds=%lld", nanoseconds);
+    NGLOG_TRACE(Kernel_SVC, "called nanoseconds={}", nanoseconds);
    // Don't attempt to yield execution if there are no available threads to run,
    // this way we avoid a useless reschedule to the idle thread.
@@ -626,111 +602,83 @@ static void SleepThread(s64 nanoseconds) {
 /// Signal process wide key atomic
 static ResultCode WaitProcessWideKeyAtomic(VAddr mutex_addr, VAddr condition_variable_addr,
                                           Handle thread_handle, s64 nano_seconds) {
-    LOG_TRACE(
+    NGLOG_TRACE(
        Kernel_SVC,
-        "called mutex_addr=%llx, condition_variable_addr=%llx, thread_handle=0x%08X, timeout=%d",
+        "called mutex_addr={:X}, condition_variable_addr={:X}, thread_handle={:#010X}, timeout={}",
        mutex_addr, condition_variable_addr, thread_handle, nano_seconds);
    SharedPtr<Thread> thread = g_handle_table.Get<Thread>(thread_handle);
    ASSERT(thread);
-    SharedPtr<Mutex> mutex = g_object_address_table.Get<Mutex>(mutex_addr);
+    CASCADE_CODE(Mutex::Release(mutex_addr));
-    if (!mutex) {
-        // Create a new mutex for the specified address if one does not already exist
-        mutex = Mutex::Create(thread, mutex_addr);
-        mutex->name = Common::StringFromFormat("mutex-%llx", mutex_addr);
-    }
-    SharedPtr<ConditionVariable> condition_variable =
+    SharedPtr<Thread> current_thread = GetCurrentThread();
-        g_object_address_table.Get<ConditionVariable>(condition_variable_addr);
+    current_thread->condvar_wait_address = condition_variable_addr;
-    if (!condition_variable) {
+    current_thread->mutex_wait_address = mutex_addr;
-        // Create a new condition_variable for the specified address if one does not already exist
+    current_thread->wait_handle = thread_handle;
-        condition_variable = ConditionVariable::Create(condition_variable_addr).Unwrap();
+    current_thread->status = THREADSTATUS_WAIT_MUTEX;
-        condition_variable->name =
+    current_thread->wakeup_callback = nullptr;
-            Common::StringFromFormat("condition-variable-%llx", condition_variable_addr);
-    }
-    if (condition_variable->mutex_addr) {
+    current_thread->WakeAfterDelay(nano_seconds);
-        // Previously created the ConditionVariable using WaitProcessWideKeyAtomic, verify
-        // everything is correct
-        ASSERT(condition_variable->mutex_addr == mutex_addr);
-    } else {
-        // Previously created the ConditionVariable using SignalProcessWideKey, set the mutex
-        // associated with it
-        condition_variable->mutex_addr = mutex_addr;
-    }
-    if (mutex->GetOwnerHandle()) {
+    // Note: Deliberately don't attempt to inherit the lock owner's priority.
-        // Release the mutex if the current thread is holding it
-        mutex->Release(thread.get());
-    }
-    auto wakeup_callback = [mutex, nano_seconds](ThreadWakeupReason reason,
+    Core::System::GetInstance().PrepareReschedule();
-                                                 SharedPtr<Thread> thread,
+    return RESULT_SUCCESS;
-                                                 SharedPtr<WaitObject> object, size_t index) {
+}
-        ASSERT(thread->status == THREADSTATUS_WAIT_SYNCH_ANY);
-        if (reason == ThreadWakeupReason::Timeout) {
+/// Signal process wide key
-            thread->SetWaitSynchronizationResult(RESULT_TIMEOUT);
+static ResultCode SignalProcessWideKey(VAddr condition_variable_addr, s32 target) {
-            return true;
+    NGLOG_TRACE(Kernel_SVC, "called, condition_variable_addr={:#X}, target={:#010X}",
-        }
+                condition_variable_addr, target);
-        ASSERT(reason == ThreadWakeupReason::Signal);
+    u32 processed = 0;
+    auto& thread_list = Core::System::GetInstance().Scheduler().GetThreadList();
-        // Now try to acquire the mutex and don't resume if it's not available.
+    for (auto& thread : thread_list) {
-        if (!mutex->ShouldWait(thread.get())) {
+        if (thread->condvar_wait_address != condition_variable_addr)
-            mutex->Acquire(thread.get());
+            continue;
-            thread->SetWaitSynchronizationResult(RESULT_SUCCESS);
-            return true;
-        }
-        if (nano_seconds == 0) {
+        // Only process up to 'target' threads, unless 'target' is -1, in which case process
-            thread->SetWaitSynchronizationResult(RESULT_TIMEOUT);
+        // them all.
-            return true;
+        if (target != -1 && processed >= target)
-        }
+            break;
-        thread->wait_objects = {mutex};
+        // If the mutex is not yet acquired, acquire it.
-        mutex->AddWaitingThread(thread);
+        u32 mutex_val = Memory::Read32(thread->mutex_wait_address);
-        thread->status = THREADSTATUS_WAIT_SYNCH_ANY;
-        // Create an event to wake the thread up after the
+        if (mutex_val == 0) {
-        // specified nanosecond delay has passed
+            // We were able to acquire the mutex, resume this thread.
-        thread->WakeAfterDelay(nano_seconds);
+            Memory::Write32(thread->mutex_wait_address, thread->wait_handle);
-        thread->wakeup_callback = DefaultThreadWakeupCallback;
+            ASSERT(thread->status == THREADSTATUS_WAIT_MUTEX);
+            thread->ResumeFromWait();
-        Core::System::GetInstance().PrepareReschedule();
+            auto lock_owner = thread->lock_owner;
+            if (lock_owner)
+                lock_owner->RemoveMutexWaiter(thread);
-        return false;
+            thread->lock_owner = nullptr;
-    };
+            thread->mutex_wait_address = 0;
-    CASCADE_CODE(
+            thread->condvar_wait_address = 0;
-        WaitSynchronization1(condition_variable, thread.get(), nano_seconds, wakeup_callback));
+            thread->wait_handle = 0;
+        } else {
+            // Couldn't acquire the mutex, block the thread.
+            Handle owner_handle = static_cast<Handle>(mutex_val & Mutex::MutexOwnerMask);
+            auto owner = g_handle_table.Get<Thread>(owner_handle);
+            ASSERT(owner);
+            ASSERT(thread->status != THREADSTATUS_RUNNING);
+            thread->status = THREADSTATUS_WAIT_MUTEX;
+            thread->wakeup_callback = nullptr;
-    return RESULT_SUCCESS;
+            // Signal that the mutex now has a waiting thread.
-}
+            Memory::Write32(thread->mutex_wait_address, mutex_val | Mutex::MutexHasWaitersFlag);
-/// Signal process wide key
+            owner->AddMutexWaiter(thread);
-static ResultCode SignalProcessWideKey(VAddr condition_variable_addr, s32 target) {
-    LOG_TRACE(Kernel_SVC, "called, condition_variable_addr=0x%llx, target=0x%08x",
-              condition_variable_addr, target);
-    // Wakeup all or one thread - Any other value is unimplemented
-    ASSERT(target == -1 || target == 1);
-    SharedPtr<ConditionVariable> condition_variable =
-        g_object_address_table.Get<ConditionVariable>(condition_variable_addr);
-    if (!condition_variable) {
-        // Create a new condition_variable for the specified address if one does not already exist
-        condition_variable = ConditionVariable::Create(condition_variable_addr).Unwrap();
-        condition_variable->name =
-            Common::StringFromFormat("condition-variable-%llx", condition_variable_addr);
-    }
-    CASCADE_CODE(condition_variable->Release(target));
+            Core::System::GetInstance().PrepareReschedule();
+        }
-    if (condition_variable->mutex_addr) {
+        ++processed;
-        // If a mutex was created for this condition_variable, wait the current thread on it
-        SharedPtr<Mutex> mutex = g_object_address_table.Get<Mutex>(condition_variable->mutex_addr);
-        return WaitSynchronization1(mutex, GetCurrentThread());
    }
    return RESULT_SUCCESS;
@@ -748,13 +696,13 @@ static u64 GetSystemTick() {
 /// Close a handle
 static ResultCode CloseHandle(Handle handle) {
-    LOG_TRACE(Kernel_SVC, "Closing handle 0x%08X", handle);
+    NGLOG_TRACE(Kernel_SVC, "Closing handle {:#010X}", handle);
    return g_handle_table.Close(handle);
 }
 /// Reset an event
 static ResultCode ResetSignal(Handle handle) {
-    LOG_WARNING(Kernel_SVC, "(STUBBED) called handle 0x%08X", handle);
+    NGLOG_WARNING(Kernel_SVC, "(STUBBED) called handle {:#010X}", handle);
    auto event = g_handle_table.Get<Event>(handle);
    ASSERT(event != nullptr);
    event->Clear();
@@ -763,29 +711,29 @@ static ResultCode ResetSignal(Handle handle) {
 /// Creates a TransferMemory object
 static ResultCode CreateTransferMemory(Handle* handle, VAddr addr, u64 size, u32 permissions) {
-    LOG_WARNING(Kernel_SVC, "(STUBBED) called addr=0x%lx, size=0x%lx, perms=%08X", addr, size,
+    NGLOG_WARNING(Kernel_SVC, "(STUBBED) called addr={:#X}, size={:#X}, perms={:010X}", addr, size,
-                permissions);
+                  permissions);
    *handle = 0;
    return RESULT_SUCCESS;
 }
 static ResultCode GetThreadCoreMask(Handle handle, u32* mask, u64* unknown) {
-    LOG_WARNING(Kernel_SVC, "(STUBBED) called, handle=0x%08X", handle);
+    NGLOG_WARNING(Kernel_SVC, "(STUBBED) called, handle={:010X}", handle);
    *mask = 0x0;
    *unknown = 0xf;
    return RESULT_SUCCESS;
 }
 static ResultCode SetThreadCoreMask(Handle handle, u32 mask, u64 unknown) {
-    LOG_WARNING(Kernel_SVC, "(STUBBED) called, handle=0x%08X, mask=0x%08X, unknown=0x%lx", handle,
+    NGLOG_WARNING(Kernel_SVC, "(STUBBED) called, handle={:#010X}, mask={:#010X}, unknown={:#X}",
-                mask, unknown);
+                  handle, mask, unknown);
    return RESULT_SUCCESS;
 }
 static ResultCode CreateSharedMemory(Handle* handle, u64 size, u32 local_permissions,
                                     u32 remote_permissions) {
-    LOG_TRACE(Kernel_SVC, "called, size=0x%llx, localPerms=0x%08x, remotePerms=0x%08x", size,
+    NGLOG_TRACE(Kernel_SVC, "called, size={:#X}, localPerms={:#010X}, remotePerms={:#010X}", size,
-              local_permissions, remote_permissions);
+                local_permissions, remote_permissions);
    auto sharedMemHandle =
        SharedMemory::Create(g_handle_table.Get<Process>(KernelHandle::CurrentProcess), size,
                             static_cast<MemoryPermission>(local_permissions),
@@ -796,7 +744,7 @@ static ResultCode CreateSharedMemory(Handle* handle, u64 size, u32 local_permiss
 }
 static ResultCode ClearEvent(Handle handle) {
-    LOG_TRACE(Kernel_SVC, "called, event=0xX", handle);
+    NGLOG_TRACE(Kernel_SVC, "called, event={:010X}", handle);
    SharedPtr<Event> evt = g_handle_table.Get<Event>(handle);
    if (evt == nullptr)
@@ -948,7 +896,7 @@ static const FunctionDef SVC_Table[] = {
 static const FunctionDef* GetSVCInfo(u32 func_num) {
    if (func_num >= std::size(SVC_Table)) {
-        LOG_ERROR(Kernel_SVC, "unknown svc=0x%02X", func_num);
+        NGLOG_ERROR(Kernel_SVC, "Unknown svc={:#04X}", func_num);
        return nullptr;
    }
    return &SVC_Table[func_num];
@@ -967,10 +915,10 @@ void CallSVC(u32 immediate) {
        if (info->func) {
            info->func();
        } else {
-            LOG_CRITICAL(Kernel_SVC, "unimplemented SVC function %s(..)", info->name);
+            NGLOG_CRITICAL(Kernel_SVC, "Unimplemented SVC function {}(..)", info->name);
        }
    } else {
-        LOG_CRITICAL(Kernel_SVC, "unknown SVC function 0x%x", immediate);
+        NGLOG_CRITICAL(Kernel_SVC, "Unknown SVC function {:#X}", immediate);
    }
 }
diff --git a/src/core/hle/kernel/thread.cpp b/src/core/hle/kernel/thread.cpp
index f3a8aa4aa..4cd57ab25 100644
--- a/src/core/hle/kernel/thread.cpp
+++ b/src/core/hle/kernel/thread.cpp
@@ -77,9 +77,6 @@ void Thread::Stop() {
    }
    wait_objects.clear();
-    // Release all the mutexes that this thread holds
-    ReleaseThreadMutexes(this);
    // Mark the TLS slot in the thread's page as free.
    u64 tls_page = (tls_address - Memory::TLS_AREA_VADDR) / Memory::PAGE_SIZE;
    u64 tls_slot =
@@ -104,9 +101,10 @@ void ExitCurrentThread() {
 * @param cycles_late The number of CPU cycles that have passed since the desired wakeup time
 */
 static void ThreadWakeupCallback(u64 thread_handle, int cycles_late) {
-    SharedPtr<Thread> thread = wakeup_callback_handle_table.Get<Thread>((Handle)thread_handle);
+    const auto proper_handle = static_cast<Handle>(thread_handle);
+    SharedPtr<Thread> thread = wakeup_callback_handle_table.Get<Thread>(proper_handle);
    if (thread == nullptr) {
-        LOG_CRITICAL(Kernel, "Callback fired for invalid thread %08X", (Handle)thread_handle);
+        NGLOG_CRITICAL(Kernel, "Callback fired for invalid thread {:08X}", proper_handle);
        return;
    }
@@ -126,6 +124,19 @@ static void ThreadWakeupCallback(u64 thread_handle, int cycles_late) {
            resume = thread->wakeup_callback(ThreadWakeupReason::Timeout, thread, nullptr, 0);
    }
+    if (thread->mutex_wait_address != 0 || thread->condvar_wait_address != 0 ||
+        thread->wait_handle) {
+        ASSERT(thread->status == THREADSTATUS_WAIT_MUTEX);
+        thread->mutex_wait_address = 0;
+        thread->condvar_wait_address = 0;
+        thread->wait_handle = 0;
+        auto lock_owner = thread->lock_owner;
+        // Threads waking up by timeout from WaitProcessWideKey do not perform priority inheritance
+        // and don't have a lock owner.
+        ASSERT(lock_owner == nullptr);
+    }
    if (resume)
        thread->ResumeFromWait();
 }
@@ -151,6 +162,7 @@ void Thread::ResumeFromWait() {
    case THREADSTATUS_WAIT_HLE_EVENT:
    case THREADSTATUS_WAIT_SLEEP:
    case THREADSTATUS_WAIT_IPC:
+    case THREADSTATUS_WAIT_MUTEX:
        break;
    case THREADSTATUS_READY:
@@ -227,19 +239,19 @@ ResultVal<SharedPtr<Thread>> Thread::Create(std::string name, VAddr entry_point,
                                            SharedPtr<Process> owner_process) {
    // Check if priority is in ranged. Lowest priority -> highest priority id.
    if (priority > THREADPRIO_LOWEST) {
-        LOG_ERROR(Kernel_SVC, "Invalid thread priority: %u", priority);
+        NGLOG_ERROR(Kernel_SVC, "Invalid thread priority: {}", priority);
        return ERR_OUT_OF_RANGE;
    }
    if (processor_id > THREADPROCESSORID_MAX) {
-        LOG_ERROR(Kernel_SVC, "Invalid processor id: %d", processor_id);
+        NGLOG_ERROR(Kernel_SVC, "Invalid processor id: {}", processor_id);
        return ERR_OUT_OF_RANGE_KERNEL;
    }
    // TODO(yuriks): Other checks, returning 0xD9001BEA
    if (!Memory::IsValidVirtualAddress(*owner_process, entry_point)) {
-        LOG_ERROR(Kernel_SVC, "(name=%s): invalid entry %016" PRIx64, name.c_str(), entry_point);
+        NGLOG_ERROR(Kernel_SVC, "(name={}): invalid entry {:016X}", name, entry_point);
        // TODO (bunnei): Find the correct error code to use here
        return ResultCode(-1);
    }
@@ -256,7 +268,9 @@ ResultVal<SharedPtr<Thread>> Thread::Create(std::string name, VAddr entry_point,
    thread->last_running_ticks = CoreTiming::GetTicks();
    thread->processor_id = processor_id;
    thread->wait_objects.clear();
-    thread->wait_address = 0;
+    thread->mutex_wait_address = 0;
+    thread->condvar_wait_address = 0;
+    thread->wait_handle = 0;
    thread->name = std::move(name);
    thread->callback_handle = wakeup_callback_handle_table.Create(thread).Unwrap();
    thread->owner_process = owner_process;
@@ -276,8 +290,8 @@ ResultVal<SharedPtr<Thread>> Thread::Create(std::string name, VAddr entry_point,
        auto& linheap_memory = memory_region->linear_heap_memory;
        if (linheap_memory->size() + Memory::PAGE_SIZE > memory_region->size) {
-            LOG_ERROR(Kernel_SVC,
+            NGLOG_ERROR(Kernel_SVC,
-                      "Not enough space in region to allocate a new TLS page for thread");
+                        "Not enough space in region to allocate a new TLS page for thread");
            return ERR_OUT_OF_MEMORY;
        }
@@ -317,17 +331,8 @@ ResultVal<SharedPtr<Thread>> Thread::Create(std::string name, VAddr entry_point,
 void Thread::SetPriority(u32 priority) {
    ASSERT_MSG(priority <= THREADPRIO_LOWEST && priority >= THREADPRIO_HIGHEST,
               "Invalid priority value.");
-    Core::System::GetInstance().Scheduler().SetThreadPriority(this, priority);
+    nominal_priority = priority;
-    nominal_priority = current_priority = priority;
+    UpdatePriority();
-}
-void Thread::UpdatePriority() {
-    u32 best_priority = nominal_priority;
-    for (auto& mutex : held_mutexes) {
-        if (mutex->priority < best_priority)
-            best_priority = mutex->priority;
-    }
-    BoostPriority(best_priority);
 }
 void Thread::BoostPriority(u32 priority) {
@@ -377,6 +382,38 @@ VAddr Thread::GetCommandBufferAddress() const {
    return GetTLSAddress() + CommandHeaderOffset;
 }
+void Thread::AddMutexWaiter(SharedPtr<Thread> thread) {
+    thread->lock_owner = this;
+    wait_mutex_threads.emplace_back(std::move(thread));
+    UpdatePriority();
+}
+void Thread::RemoveMutexWaiter(SharedPtr<Thread> thread) {
+    boost::remove_erase(wait_mutex_threads, thread);
+    thread->lock_owner = nullptr;
+    UpdatePriority();
+}
+void Thread::UpdatePriority() {
+    // Find the highest priority among all the threads that are waiting for this thread's lock
+    u32 new_priority = nominal_priority;
+    for (const auto& thread : wait_mutex_threads) {
+        if (thread->nominal_priority < new_priority)
+            new_priority = thread->nominal_priority;
+    }
+    if (new_priority == current_priority)
+        return;
+    Core::System::GetInstance().Scheduler().SetThreadPriority(this, new_priority);
+    current_priority = new_priority;
+    // Recursively update the priority of the thread that depends on the priority of this one.
+    if (lock_owner)
+        lock_owner->UpdatePriority();
+}
 ////////////////////////////////////////////////////////////////////////////////////////////////////
 /**
diff --git a/src/core/hle/kernel/thread.h b/src/core/hle/kernel/thread.h
index dbf47e269..e0a3c0934 100644
--- a/src/core/hle/kernel/thread.h
+++ b/src/core/hle/kernel/thread.h
@@ -18,7 +18,7 @@
 enum ThreadPriority : u32 {
    THREADPRIO_HIGHEST = 0,       ///< Highest thread priority
    THREADPRIO_USERLAND_MAX = 24, ///< Highest thread priority for userland apps
-    THREADPRIO_DEFAULT = 48,      ///< Default thread priority for userland apps
+    THREADPRIO_DEFAULT = 44,      ///< Default thread priority for userland apps
    THREADPRIO_LOWEST = 63,       ///< Lowest thread priority
 };
@@ -43,6 +43,7 @@ enum ThreadStatus {
    THREADSTATUS_WAIT_IPC,       ///< Waiting for the reply from an IPC request
    THREADSTATUS_WAIT_SYNCH_ANY, ///< Waiting due to WaitSynch1 or WaitSynchN with wait_all = false
    THREADSTATUS_WAIT_SYNCH_ALL, ///< Waiting due to WaitSynchronizationN with wait_all = true
+    THREADSTATUS_WAIT_MUTEX,     ///< Waiting due to an ArbitrateLock/WaitProcessWideKey svc
    THREADSTATUS_DORMANT,        ///< Created but not yet made ready
    THREADSTATUS_DEAD            ///< Run to completion, or forcefully terminated
 };
@@ -54,7 +55,6 @@ enum class ThreadWakeupReason {
 namespace Kernel {
-class Mutex;
 class Process;
 class Thread final : public WaitObject {
@@ -104,17 +104,20 @@ public:
    void SetPriority(u32 priority);
    /**
-     * Boost's a thread's priority to the best priority among the thread's held mutexes.
-     * This prevents priority inversion via priority inheritance.
-     */
-    void UpdatePriority();
-    /**
     * Temporarily boosts the thread's priority until the next time it is scheduled
     * @param priority The new priority
     */
    void BoostPriority(u32 priority);
+    /// Adds a thread to the list of threads that are waiting for a lock held by this thread.
+    void AddMutexWaiter(SharedPtr<Thread> thread);
+    /// Removes a thread from the list of threads that are waiting for a lock held by this thread.
+    void RemoveMutexWaiter(SharedPtr<Thread> thread);
+    /// Recalculates the current priority taking into account priority inheritance.
+    void UpdatePriority();
    /**
     * Gets the thread's thread ID
     * @return The thread's ID
@@ -205,19 +208,22 @@ public:
    VAddr tls_address; ///< Virtual address of the Thread Local Storage of the thread
-    /// Mutexes currently held by this thread, which will be released when it exits.
-    boost::container::flat_set<SharedPtr<Mutex>> held_mutexes;
-    /// Mutexes that this thread is currently waiting for.
-    boost::container::flat_set<SharedPtr<Mutex>> pending_mutexes;
    SharedPtr<Process> owner_process; ///< Process that owns this thread
    /// Objects that the thread is waiting on, in the same order as they were
    // passed to WaitSynchronization1/N.
    std::vector<SharedPtr<WaitObject>> wait_objects;
-    VAddr wait_address; ///< If waiting on an AddressArbiter, this is the arbitration address
+    /// List of threads that are waiting for a mutex that is held by this thread.
+    std::vector<SharedPtr<Thread>> wait_mutex_threads;
+    /// Thread that owns the lock that this thread is waiting for.
+    SharedPtr<Thread> lock_owner;
+    // If waiting on a ConditionVariable, this is the ConditionVariable  address
+    VAddr condvar_wait_address;
+    VAddr mutex_wait_address; ///< If waiting on a Mutex, this is the mutex address
+    Handle wait_handle;       ///< The handle used to wait for the mutex.
    std::string name;
diff --git a/src/core/hle/kernel/timer.cpp b/src/core/hle/kernel/timer.cpp
index 8da745634..ad58bf043 100644
--- a/src/core/hle/kernel/timer.cpp
+++ b/src/core/hle/kernel/timer.cpp
@@ -77,7 +77,7 @@ void Timer::WakeupAllWaitingThreads() {
 }
 void Timer::Signal(int cycles_late) {
-    LOG_TRACE(Kernel, "Timer %u fired", GetObjectId());
+    NGLOG_TRACE(Kernel, "Timer {} fired", GetObjectId());
    signaled = true;
@@ -97,7 +97,7 @@ static void TimerCallback(u64 timer_handle, int cycles_late) {
        timer_callback_handle_table.Get<Timer>(static_cast<Handle>(timer_handle));
    if (timer == nullptr) {
-        LOG_CRITICAL(Kernel, "Callback fired for invalid timer %08" PRIx64, timer_handle);
+        NGLOG_CRITICAL(Kernel, "Callback fired for invalid timer {:016X}", timer_handle);
        return;
    }
diff --git a/src/core/hle/kernel/vm_manager.cpp b/src/core/hle/kernel/vm_manager.cpp
index acd65ee68..eb2e35eed 100644
--- a/src/core/hle/kernel/vm_manager.cpp
+++ b/src/core/hle/kernel/vm_manager.cpp
@@ -379,22 +379,22 @@ void VMManager::UpdatePageTableForVMA(const VirtualMemoryArea& vma) {
 }
 u64 VMManager::GetTotalMemoryUsage() {
-    LOG_WARNING(Kernel, "(STUBBED) called");
+    NGLOG_WARNING(Kernel, "(STUBBED) called");
    return 0xF8000000;
 }
 u64 VMManager::GetTotalHeapUsage() {
-    LOG_WARNING(Kernel, "(STUBBED) called");
+    NGLOG_WARNING(Kernel, "(STUBBED) called");
    return 0x0;
 }
 VAddr VMManager::GetAddressSpaceBaseAddr() {
-    LOG_WARNING(Kernel, "(STUBBED) called");
+    NGLOG_WARNING(Kernel, "(STUBBED) called");
    return 0x8000000;
 }
 u64 VMManager::GetAddressSpaceSize() {
-    LOG_WARNING(Kernel, "(STUBBED) called");
+    NGLOG_WARNING(Kernel, "(STUBBED) called");
    return MAX_ADDRESS;
 }
diff --git a/src/core/hle/service/acc/acc.cpp b/src/core/hle/service/acc/acc.cpp
index 6bafb2dce..f2fffa760 100644
--- a/src/core/hle/service/acc/acc.cpp
+++ b/src/core/hle/service/acc/acc.cpp
@@ -47,7 +47,7 @@ public:
 private:
    void GetBase(Kernel::HLERequestContext& ctx) {
-        LOG_WARNING(Service_ACC, "(STUBBED) called");
+        NGLOG_WARNING(Service_ACC, "(STUBBED) called");
        ProfileBase profile_base{};
        IPC::ResponseBuilder rb{ctx, 16};
        rb.Push(RESULT_SUCCESS);
@@ -72,14 +72,14 @@ public:
 private:
    void CheckAvailability(Kernel::HLERequestContext& ctx) {
-        LOG_WARNING(Service_ACC, "(STUBBED) called");
+        NGLOG_WARNING(Service_ACC, "(STUBBED) called");
        IPC::ResponseBuilder rb{ctx, 3};
        rb.Push(RESULT_SUCCESS);
        rb.Push(true); // TODO: Check when this is supposed to return true and when not
    }
    void GetAccountId(Kernel::HLERequestContext& ctx) {
-        LOG_WARNING(Service_ACC, "(STUBBED) called");
+        NGLOG_WARNING(Service_ACC, "(STUBBED) called");
        IPC::ResponseBuilder rb{ctx, 4};
        rb.Push(RESULT_SUCCESS);
        rb.Push<u64>(0x12345678ABCDEF);
@@ -87,14 +87,14 @@ private:
 };
 void Module::Interface::GetUserExistence(Kernel::HLERequestContext& ctx) {
-    LOG_WARNING(Service_ACC, "(STUBBED) called");
+    NGLOG_WARNING(Service_ACC, "(STUBBED) called");
    IPC::ResponseBuilder rb{ctx, 3};
    rb.Push(RESULT_SUCCESS);
    rb.Push(true); // TODO: Check when this is supposed to return true and when not
 }
 void Module::Interface::ListAllUsers(Kernel::HLERequestContext& ctx) {
-    LOG_WARNING(Service_ACC, "(STUBBED) called");
+    NGLOG_WARNING(Service_ACC, "(STUBBED) called");
    constexpr std::array<u128, 10> user_ids{DEFAULT_USER_ID};
    ctx.WriteBuffer(user_ids.data(), user_ids.size());
    IPC::ResponseBuilder rb{ctx, 2};
@@ -102,7 +102,7 @@ void Module::Interface::ListAllUsers(Kernel::HLERequestContext& ctx) {
 }
 void Module::Interface::ListOpenUsers(Kernel::HLERequestContext& ctx) {
-    LOG_WARNING(Service_ACC, "(STUBBED) called");
+    NGLOG_WARNING(Service_ACC, "(STUBBED) called");
    constexpr std::array<u128, 10> user_ids{DEFAULT_USER_ID};
    ctx.WriteBuffer(user_ids.data(), user_ids.size());
    IPC::ResponseBuilder rb{ctx, 2};
@@ -113,11 +113,11 @@ void Module::Interface::GetProfile(Kernel::HLERequestContext& ctx) {
    IPC::ResponseBuilder rb{ctx, 2, 0, 1};
    rb.Push(RESULT_SUCCESS);
    rb.PushIpcInterface<IProfile>();
-    LOG_DEBUG(Service_ACC, "called");
+    NGLOG_DEBUG(Service_ACC, "called");
 }
 void Module::Interface::InitializeApplicationInfo(Kernel::HLERequestContext& ctx) {
-    LOG_WARNING(Service_ACC, "(STUBBED) called");
+    NGLOG_WARNING(Service_ACC, "(STUBBED) called");
    IPC::ResponseBuilder rb{ctx, 2};
    rb.Push(RESULT_SUCCESS);
 }
@@ -126,11 +126,11 @@ void Module::Interface::GetBaasAccountManagerForApplication(Kernel::HLERequestCo
    IPC::ResponseBuilder rb{ctx, 2, 0, 1};
    rb.Push(RESULT_SUCCESS);
    rb.PushIpcInterface<IManagerForApplication>();
-    LOG_DEBUG(Service_ACC, "called");
+    NGLOG_DEBUG(Service_ACC, "called");
 }
 void Module::Interface::GetLastOpenedUser(Kernel::HLERequestContext& ctx) {
-    LOG_WARNING(Service_ACC, "(STUBBED) called");
+    NGLOG_WARNING(Service_ACC, "(STUBBED) called");
    IPC::ResponseBuilder rb{ctx, 6};
    rb.Push(RESULT_SUCCESS);
    rb.PushRaw(DEFAULT_USER_ID);
diff --git a/src/core/hle/service/am/am.cpp b/src/core/hle/service/am/am.cpp
index f41a59afe..19fadcb8e 100644
--- a/src/core/hle/service/am/am.cpp
+++ b/src/core/hle/service/am/am.cpp
@@ -28,14 +28,14 @@ IWindowController::IWindowController() : ServiceFramework("IWindowController") {
 }
 void IWindowController::GetAppletResourceUserId(Kernel::HLERequestContext& ctx) {
-    LOG_WARNING(Service_AM, "(STUBBED) called");
+    NGLOG_WARNING(Service_AM, "(STUBBED) called");
    IPC::ResponseBuilder rb{ctx, 4};
    rb.Push(RESULT_SUCCESS);
    rb.Push<u64>(0);
 }
 void IWindowController::AcquireForegroundRights(Kernel::HLERequestContext& ctx) {
-    LOG_WARNING(Service_AM, "(STUBBED) called");
+    NGLOG_WARNING(Service_AM, "(STUBBED) called");
    IPC::ResponseBuilder rb{ctx, 2};
    rb.Push(RESULT_SUCCESS);
 }
@@ -54,20 +54,20 @@ IAudioController::IAudioController() : ServiceFramework("IAudioController") {
 }
 void IAudioController::SetExpectedMasterVolume(Kernel::HLERequestContext& ctx) {
-    LOG_WARNING(Service_AM, "(STUBBED) called");
+    NGLOG_WARNING(Service_AM, "(STUBBED) called");
    IPC::ResponseBuilder rb{ctx, 2};
    rb.Push(RESULT_SUCCESS);
 }
 void IAudioController::GetMainAppletExpectedMasterVolume(Kernel::HLERequestContext& ctx) {
-    LOG_WARNING(Service_AM, "(STUBBED) called");
+    NGLOG_WARNING(Service_AM, "(STUBBED) called");
    IPC::ResponseBuilder rb{ctx, 3};
    rb.Push(RESULT_SUCCESS);
    rb.Push(volume);
 }
 void IAudioController::GetLibraryAppletExpectedMasterVolume(Kernel::HLERequestContext& ctx) {
-    LOG_WARNING(Service_AM, "(STUBBED) called");
+    NGLOG_WARNING(Service_AM, "(STUBBED) called");
    IPC::ResponseBuilder rb{ctx, 3};
    rb.Push(RESULT_SUCCESS);
    rb.Push(volume);
@@ -139,14 +139,14 @@ void ISelfController::SetFocusHandlingMode(Kernel::HLERequestContext& ctx) {
    IPC::ResponseBuilder rb{ctx, 2};
    rb.Push(RESULT_SUCCESS);
-    LOG_WARNING(Service_AM, "(STUBBED) called");
+    NGLOG_WARNING(Service_AM, "(STUBBED) called");
 }
 void ISelfController::SetRestartMessageEnabled(Kernel::HLERequestContext& ctx) {
    IPC::ResponseBuilder rb{ctx, 2};
    rb.Push(RESULT_SUCCESS);
-    LOG_WARNING(Service_AM, "(STUBBED) called");
+    NGLOG_WARNING(Service_AM, "(STUBBED) called");
 }
 void ISelfController::SetPerformanceModeChangedNotification(Kernel::HLERequestContext& ctx) {
@@ -157,14 +157,14 @@ void ISelfController::SetPerformanceModeChangedNotification(Kernel::HLERequestCo
    IPC::ResponseBuilder rb{ctx, 2};
    rb.Push(RESULT_SUCCESS);
-    LOG_WARNING(Service_AM, "(STUBBED) called flag=%u", static_cast<u32>(flag));
+    NGLOG_WARNING(Service_AM, "(STUBBED) called flag={}", flag);
 }
 void ISelfController::SetScreenShotPermission(Kernel::HLERequestContext& ctx) {
    IPC::ResponseBuilder rb{ctx, 2};
    rb.Push(RESULT_SUCCESS);
-    LOG_WARNING(Service_AM, "(STUBBED) called");
+    NGLOG_WARNING(Service_AM, "(STUBBED) called");
 }
 void ISelfController::SetOperationModeChangedNotification(Kernel::HLERequestContext& ctx) {
@@ -175,7 +175,7 @@ void ISelfController::SetOperationModeChangedNotification(Kernel::HLERequestCont
    IPC::ResponseBuilder rb{ctx, 2};
    rb.Push(RESULT_SUCCESS);
-    LOG_WARNING(Service_AM, "(STUBBED) called flag=%u", static_cast<u32>(flag));
+    NGLOG_WARNING(Service_AM, "(STUBBED) called flag={}", flag);
 }
 void ISelfController::SetOutOfFocusSuspendingEnabled(Kernel::HLERequestContext& ctx) {
@@ -188,21 +188,21 @@ void ISelfController::SetOutOfFocusSuspendingEnabled(Kernel::HLERequestContext&
    IPC::ResponseBuilder rb{ctx, 2};
    rb.Push(RESULT_SUCCESS);
-    LOG_WARNING(Service_AM, "(STUBBED) called enabled=%u", static_cast<u32>(enabled));
+    NGLOG_WARNING(Service_AM, "(STUBBED) called enabled={}", enabled);
 }
 void ISelfController::LockExit(Kernel::HLERequestContext& ctx) {
    IPC::ResponseBuilder rb{ctx, 2};
    rb.Push(RESULT_SUCCESS);
-    LOG_WARNING(Service_AM, "(STUBBED) called");
+    NGLOG_WARNING(Service_AM, "(STUBBED) called");
 }
 void ISelfController::UnlockExit(Kernel::HLERequestContext& ctx) {
    IPC::ResponseBuilder rb{ctx, 2};
    rb.Push(RESULT_SUCCESS);
-    LOG_WARNING(Service_AM, "(STUBBED) called");
+    NGLOG_WARNING(Service_AM, "(STUBBED) called");
 }
 void ISelfController::GetLibraryAppletLaunchableEvent(Kernel::HLERequestContext& ctx) {
@@ -212,7 +212,7 @@ void ISelfController::GetLibraryAppletLaunchableEvent(Kernel::HLERequestContext&
    rb.Push(RESULT_SUCCESS);
    rb.PushCopyObjects(launchable_event);
-    LOG_WARNING(Service_AM, "(STUBBED) called");
+    NGLOG_WARNING(Service_AM, "(STUBBED) called");
 }
 void ISelfController::CreateManagedDisplayLayer(Kernel::HLERequestContext& ctx) {
@@ -225,7 +225,7 @@ void ISelfController::CreateManagedDisplayLayer(Kernel::HLERequestContext& ctx)
    rb.Push(RESULT_SUCCESS);
    rb.Push(layer_id);
-    LOG_WARNING(Service_AM, "(STUBBED) called");
+    NGLOG_WARNING(Service_AM, "(STUBBED) called");
 }
 ICommonStateGetter::ICommonStateGetter() : ServiceFramework("ICommonStateGetter") {
@@ -269,7 +269,7 @@ void ICommonStateGetter::GetEventHandle(Kernel::HLERequestContext& ctx) {
    rb.Push(RESULT_SUCCESS);
    rb.PushCopyObjects(event);
-    LOG_WARNING(Service_AM, "(STUBBED) called");
+    NGLOG_WARNING(Service_AM, "(STUBBED) called");
 }
 void ICommonStateGetter::ReceiveMessage(Kernel::HLERequestContext& ctx) {
@@ -277,7 +277,7 @@ void ICommonStateGetter::ReceiveMessage(Kernel::HLERequestContext& ctx) {
    rb.Push(RESULT_SUCCESS);
    rb.Push<u32>(15);
-    LOG_WARNING(Service_AM, "(STUBBED) called");
+    NGLOG_WARNING(Service_AM, "(STUBBED) called");
 }
 void ICommonStateGetter::GetCurrentFocusState(Kernel::HLERequestContext& ctx) {
@@ -285,7 +285,7 @@ void ICommonStateGetter::GetCurrentFocusState(Kernel::HLERequestContext& ctx) {
    rb.Push(RESULT_SUCCESS);
    rb.Push(static_cast<u8>(FocusState::InFocus));
-    LOG_WARNING(Service_AM, "(STUBBED) called");
+    NGLOG_WARNING(Service_AM, "(STUBBED) called");
 }
 void ICommonStateGetter::GetOperationMode(Kernel::HLERequestContext& ctx) {
@@ -294,7 +294,7 @@ void ICommonStateGetter::GetOperationMode(Kernel::HLERequestContext& ctx) {
    rb.Push(RESULT_SUCCESS);
    rb.Push(static_cast<u8>(use_docked_mode ? OperationMode::Docked : OperationMode::Handheld));
-    LOG_WARNING(Service_AM, "(STUBBED) called");
+    NGLOG_WARNING(Service_AM, "(STUBBED) called");
 }
 void ICommonStateGetter::GetPerformanceMode(Kernel::HLERequestContext& ctx) {
@@ -304,7 +304,7 @@ void ICommonStateGetter::GetPerformanceMode(Kernel::HLERequestContext& ctx) {
    rb.Push(static_cast<u32>(use_docked_mode ? APM::PerformanceMode::Docked
                                             : APM::PerformanceMode::Handheld));
-    LOG_WARNING(Service_AM, "(STUBBED) called");
+    NGLOG_WARNING(Service_AM, "(STUBBED) called");
 }
 class ILibraryAppletAccessor final : public ServiceFramework<ILibraryAppletAccessor> {
@@ -344,7 +344,7 @@ private:
        rb.Push(RESULT_SUCCESS);
        rb.PushCopyObjects(state_changed_event);
-        LOG_WARNING(Service_AM, "(STUBBED) called");
+        NGLOG_WARNING(Service_AM, "(STUBBED) called");
    }
    Kernel::SharedPtr<Kernel::Event> state_changed_event;
@@ -368,7 +368,7 @@ void ILibraryAppletCreator::CreateLibraryApplet(Kernel::HLERequestContext& ctx)
    rb.Push(RESULT_SUCCESS);
    rb.PushIpcInterface<AM::ILibraryAppletAccessor>();
-    LOG_DEBUG(Service_AM, "called");
+    NGLOG_DEBUG(Service_AM, "called");
 }
 class IStorageAccessor final : public ServiceFramework<IStorageAccessor> {
@@ -392,7 +392,7 @@ private:
        rb.Push(RESULT_SUCCESS);
        rb.Push(static_cast<u64>(buffer.size()));
-        LOG_DEBUG(Service_AM, "called");
+        NGLOG_DEBUG(Service_AM, "called");
    }
    void Read(Kernel::HLERequestContext& ctx) {
@@ -410,7 +410,7 @@ private:
        rb.Push(RESULT_SUCCESS);
-        LOG_DEBUG(Service_AM, "called");
+        NGLOG_DEBUG(Service_AM, "called");
    }
 };
@@ -434,7 +434,7 @@ private:
        rb.Push(RESULT_SUCCESS);
        rb.PushIpcInterface<AM::IStorageAccessor>(buffer);
-        LOG_DEBUG(Service_AM, "called");
+        NGLOG_DEBUG(Service_AM, "called");
    }
 };
@@ -498,14 +498,14 @@ void IApplicationFunctions::PopLaunchParameter(Kernel::HLERequestContext& ctx) {
    rb.Push(RESULT_SUCCESS);
    rb.PushIpcInterface<AM::IStorage>(buffer);
-    LOG_DEBUG(Service_AM, "called");
+    NGLOG_DEBUG(Service_AM, "called");
 }
 void IApplicationFunctions::EnsureSaveData(Kernel::HLERequestContext& ctx) {
    IPC::RequestParser rp{ctx};
    u128 uid = rp.PopRaw<u128>();
-    LOG_WARNING(Service, "(STUBBED) called uid = %016" PRIX64 "%016" PRIX64, uid[1], uid[0]);
+    NGLOG_WARNING(Service, "(STUBBED) called uid = {:016X}{:016X}", uid[1], uid[0]);
    IPC::ResponseBuilder rb{ctx, 4};
@@ -533,27 +533,27 @@ void IApplicationFunctions::SetTerminateResult(Kernel::HLERequestContext& ctx) {
    IPC::ResponseBuilder rb{ctx, 2};
    rb.Push(RESULT_SUCCESS);
-    LOG_WARNING(Service_AM, "(STUBBED) called, result=0x%08X", result);
+    NGLOG_WARNING(Service_AM, "(STUBBED) called, result={:#010}", result);
 }
 void IApplicationFunctions::GetDesiredLanguage(Kernel::HLERequestContext& ctx) {
    IPC::ResponseBuilder rb{ctx, 4};
    rb.Push(RESULT_SUCCESS);
    rb.Push<u64>(SystemLanguage::English);
-    LOG_WARNING(Service_AM, "(STUBBED) called");
+    NGLOG_WARNING(Service_AM, "(STUBBED) called");
 }
 void IApplicationFunctions::InitializeGamePlayRecording(Kernel::HLERequestContext& ctx) {
    IPC::ResponseBuilder rb{ctx, 2};
    rb.Push(RESULT_SUCCESS);
-    LOG_WARNING(Service_AM, "(STUBBED) called");
+    NGLOG_WARNING(Service_AM, "(STUBBED) called");
 }
 void IApplicationFunctions::SetGamePlayRecordingState(Kernel::HLERequestContext& ctx) {
    IPC::ResponseBuilder rb{ctx, 2};
    rb.Push(RESULT_SUCCESS);
-    LOG_WARNING(Service_AM, "(STUBBED) called");
+    NGLOG_WARNING(Service_AM, "(STUBBED) called");
 }
 void IApplicationFunctions::NotifyRunning(Kernel::HLERequestContext& ctx) {
@@ -561,7 +561,7 @@ void IApplicationFunctions::NotifyRunning(Kernel::HLERequestContext& ctx) {
    rb.Push(RESULT_SUCCESS);
    rb.Push<u8>(0); // Unknown, seems to be ignored by official processes
-    LOG_WARNING(Service_AM, "(STUBBED) called");
+    NGLOG_WARNING(Service_AM, "(STUBBED) called");
 }
 void InstallInterfaces(SM::ServiceManager& service_manager,
diff --git a/src/core/hle/service/am/applet_ae.cpp b/src/core/hle/service/am/applet_ae.cpp
index 4f0698a8a..8951980cf 100644
--- a/src/core/hle/service/am/applet_ae.cpp
+++ b/src/core/hle/service/am/applet_ae.cpp
@@ -33,56 +33,56 @@ private:
        IPC::ResponseBuilder rb{ctx, 2, 0, 1};
        rb.Push(RESULT_SUCCESS);
        rb.PushIpcInterface<ICommonStateGetter>();
-        LOG_DEBUG(Service_AM, "called");
+        NGLOG_DEBUG(Service_AM, "called");
    }
    void GetSelfController(Kernel::HLERequestContext& ctx) {
        IPC::ResponseBuilder rb{ctx, 2, 0, 1};
        rb.Push(RESULT_SUCCESS);
        rb.PushIpcInterface<ISelfController>(nvflinger);
-        LOG_DEBUG(Service_AM, "called");
+        NGLOG_DEBUG(Service_AM, "called");
    }
    void GetWindowController(Kernel::HLERequestContext& ctx) {
        IPC::ResponseBuilder rb{ctx, 2, 0, 1};
        rb.Push(RESULT_SUCCESS);
        rb.PushIpcInterface<IWindowController>();
-        LOG_DEBUG(Service_AM, "called");
+        NGLOG_DEBUG(Service_AM, "called");
    }
    void GetAudioController(Kernel::HLERequestContext& ctx) {
        IPC::ResponseBuilder rb{ctx, 2, 0, 1};
        rb.Push(RESULT_SUCCESS);
        rb.PushIpcInterface<IAudioController>();
-        LOG_DEBUG(Service_AM, "called");
+        NGLOG_DEBUG(Service_AM, "called");
    }
    void GetDisplayController(Kernel::HLERequestContext& ctx) {
        IPC::ResponseBuilder rb{ctx, 2, 0, 1};
        rb.Push(RESULT_SUCCESS);
        rb.PushIpcInterface<IDisplayController>();
-        LOG_DEBUG(Service_AM, "called");
+        NGLOG_DEBUG(Service_AM, "called");
    }
    void GetDebugFunctions(Kernel::HLERequestContext& ctx) {
        IPC::ResponseBuilder rb{ctx, 2, 0, 1};
        rb.Push(RESULT_SUCCESS);
        rb.PushIpcInterface<IDebugFunctions>();
-        LOG_DEBUG(Service_AM, "called");
+        NGLOG_DEBUG(Service_AM, "called");
    }
    void GetLibraryAppletCreator(Kernel::HLERequestContext& ctx) {
        IPC::ResponseBuilder rb{ctx, 2, 0, 1};
        rb.Push(RESULT_SUCCESS);
        rb.PushIpcInterface<ILibraryAppletCreator>();
-        LOG_DEBUG(Service_AM, "called");
+        NGLOG_DEBUG(Service_AM, "called");
    }
    void GetApplicationFunctions(Kernel::HLERequestContext& ctx) {
        IPC::ResponseBuilder rb{ctx, 2, 0, 1};
        rb.Push(RESULT_SUCCESS);
        rb.PushIpcInterface<IApplicationFunctions>();
-        LOG_DEBUG(Service_AM, "called");
+        NGLOG_DEBUG(Service_AM, "called");
    }
    std::shared_ptr<NVFlinger::NVFlinger> nvflinger;
@@ -92,7 +92,7 @@ void AppletAE::OpenLibraryAppletProxyOld(Kernel::HLERequestContext& ctx) {
    IPC::ResponseBuilder rb{ctx, 2, 0, 1};
    rb.Push(RESULT_SUCCESS);
    rb.PushIpcInterface<ILibraryAppletProxy>(nvflinger);
-    LOG_DEBUG(Service_AM, "called");
+    NGLOG_DEBUG(Service_AM, "called");
 }
 AppletAE::AppletAE(std::shared_ptr<NVFlinger::NVFlinger> nvflinger)
diff --git a/src/core/hle/service/am/applet_oe.cpp b/src/core/hle/service/am/applet_oe.cpp
index 674b4d753..68388bf5e 100644
--- a/src/core/hle/service/am/applet_oe.cpp
+++ b/src/core/hle/service/am/applet_oe.cpp
@@ -33,56 +33,56 @@ private:
        IPC::ResponseBuilder rb{ctx, 2, 0, 1};
        rb.Push(RESULT_SUCCESS);
        rb.PushIpcInterface<IAudioController>();
-        LOG_DEBUG(Service_AM, "called");
+        NGLOG_DEBUG(Service_AM, "called");
    }
    void GetDisplayController(Kernel::HLERequestContext& ctx) {
        IPC::ResponseBuilder rb{ctx, 2, 0, 1};
        rb.Push(RESULT_SUCCESS);
        rb.PushIpcInterface<IDisplayController>();
-        LOG_DEBUG(Service_AM, "called");
+        NGLOG_DEBUG(Service_AM, "called");
    }
    void GetDebugFunctions(Kernel::HLERequestContext& ctx) {
        IPC::ResponseBuilder rb{ctx, 2, 0, 1};
        rb.Push(RESULT_SUCCESS);
        rb.PushIpcInterface<IDebugFunctions>();
-        LOG_DEBUG(Service_AM, "called");
+        NGLOG_DEBUG(Service_AM, "called");
    }
    void GetWindowController(Kernel::HLERequestContext& ctx) {
        IPC::ResponseBuilder rb{ctx, 2, 0, 1};
        rb.Push(RESULT_SUCCESS);
        rb.PushIpcInterface<IWindowController>();
-        LOG_DEBUG(Service_AM, "called");
+        NGLOG_DEBUG(Service_AM, "called");
    }
    void GetSelfController(Kernel::HLERequestContext& ctx) {
        IPC::ResponseBuilder rb{ctx, 2, 0, 1};
        rb.Push(RESULT_SUCCESS);
        rb.PushIpcInterface<ISelfController>(nvflinger);
-        LOG_DEBUG(Service_AM, "called");
+        NGLOG_DEBUG(Service_AM, "called");
    }
    void GetCommonStateGetter(Kernel::HLERequestContext& ctx) {
        IPC::ResponseBuilder rb{ctx, 2, 0, 1};
        rb.Push(RESULT_SUCCESS);
        rb.PushIpcInterface<ICommonStateGetter>();
-        LOG_DEBUG(Service_AM, "called");
+        NGLOG_DEBUG(Service_AM, "called");
    }
    void GetLibraryAppletCreator(Kernel::HLERequestContext& ctx) {
        IPC::ResponseBuilder rb{ctx, 2, 0, 1};
        rb.Push(RESULT_SUCCESS);
        rb.PushIpcInterface<ILibraryAppletCreator>();
-        LOG_DEBUG(Service_AM, "called");
+        NGLOG_DEBUG(Service_AM, "called");
    }
    void GetApplicationFunctions(Kernel::HLERequestContext& ctx) {
        IPC::ResponseBuilder rb{ctx, 2, 0, 1};
        rb.Push(RESULT_SUCCESS);
        rb.PushIpcInterface<IApplicationFunctions>();
-        LOG_DEBUG(Service_AM, "called");
+        NGLOG_DEBUG(Service_AM, "called");
    }
    std::shared_ptr<NVFlinger::NVFlinger> nvflinger;
@@ -92,7 +92,7 @@ void AppletOE::OpenApplicationProxy(Kernel::HLERequestContext& ctx) {
    IPC::ResponseBuilder rb{ctx, 2, 0, 1};
    rb.Push(RESULT_SUCCESS);
    rb.PushIpcInterface<IApplicationProxy>(nvflinger);
-    LOG_DEBUG(Service_AM, "called");
+    NGLOG_DEBUG(Service_AM, "called");
 }
 AppletOE::AppletOE(std::shared_ptr<NVFlinger::NVFlinger> nvflinger)
diff --git a/src/core/hle/service/aoc/aoc_u.cpp b/src/core/hle/service/aoc/aoc_u.cpp
index 6e7438580..5b6dfb48f 100644
--- a/src/core/hle/service/aoc/aoc_u.cpp
+++ b/src/core/hle/service/aoc/aoc_u.cpp
@@ -27,14 +27,14 @@ void AOC_U::CountAddOnContent(Kernel::HLERequestContext& ctx) {
    IPC::ResponseBuilder rb{ctx, 4};
    rb.Push(RESULT_SUCCESS);
    rb.Push<u64>(0);
-    LOG_WARNING(Service_AOC, "(STUBBED) called");
+    NGLOG_WARNING(Service_AOC, "(STUBBED) called");
 }
 void AOC_U::ListAddOnContent(Kernel::HLERequestContext& ctx) {
    IPC::ResponseBuilder rb{ctx, 4};
    rb.Push(RESULT_SUCCESS);
    rb.Push<u64>(0);
-    LOG_WARNING(Service_AOC, "(STUBBED) called");
+    NGLOG_WARNING(Service_AOC, "(STUBBED) called");
 }
 void InstallInterfaces(SM::ServiceManager& service_manager) {
diff --git a/src/core/hle/service/apm/interface.cpp b/src/core/hle/service/apm/interface.cpp
index 4e11f3f14..3a03188ce 100644
--- a/src/core/hle/service/apm/interface.cpp
+++ b/src/core/hle/service/apm/interface.cpp
@@ -29,8 +29,8 @@ private:
        IPC::ResponseBuilder rb{ctx, 2};
        rb.Push(RESULT_SUCCESS);
-        LOG_WARNING(Service_APM, "(STUBBED) called mode=%u config=%u", static_cast<u32>(mode),
+        NGLOG_WARNING(Service_APM, "(STUBBED) called mode={} config={}", static_cast<u32>(mode),
-                    config);
+                      config);
    }
    void GetPerformanceConfiguration(Kernel::HLERequestContext& ctx) {
@@ -42,7 +42,7 @@ private:
        rb.Push(RESULT_SUCCESS);
        rb.Push<u32>(0); // Performance configuration
-        LOG_WARNING(Service_APM, "(STUBBED) called mode=%u", static_cast<u32>(mode));
+        NGLOG_WARNING(Service_APM, "(STUBBED) called mode={}", static_cast<u32>(mode));
    }
 };
diff --git a/src/core/hle/service/audio/audout_u.cpp b/src/core/hle/service/audio/audout_u.cpp
index 2d7f8cb04..6297dc450 100644
--- a/src/core/hle/service/audio/audout_u.cpp
+++ b/src/core/hle/service/audio/audout_u.cpp
@@ -60,14 +60,14 @@ public:
 private:
    void GetAudioOutState(Kernel::HLERequestContext& ctx) {
-        LOG_DEBUG(Service_Audio, "called");
+        NGLOG_DEBUG(Service_Audio, "called");
        IPC::ResponseBuilder rb{ctx, 3};
        rb.Push(RESULT_SUCCESS);
        rb.Push(static_cast<u32>(audio_out_state));
    }
    void StartAudioOut(Kernel::HLERequestContext& ctx) {
-        LOG_WARNING(Service_Audio, "(STUBBED) called");
+        NGLOG_WARNING(Service_Audio, "(STUBBED) called");
        // Start audio
        audio_out_state = AudioState::Started;
@@ -77,7 +77,7 @@ private:
    }
    void StopAudioOut(Kernel::HLERequestContext& ctx) {
-        LOG_WARNING(Service_Audio, "(STUBBED) called");
+        NGLOG_WARNING(Service_Audio, "(STUBBED) called");
        // Stop audio
        audio_out_state = AudioState::Stopped;
@@ -89,7 +89,7 @@ private:
    }
    void RegisterBufferEvent(Kernel::HLERequestContext& ctx) {
-        LOG_WARNING(Service_Audio, "(STUBBED) called");
+        NGLOG_WARNING(Service_Audio, "(STUBBED) called");
        IPC::ResponseBuilder rb{ctx, 2, 1};
        rb.Push(RESULT_SUCCESS);
@@ -97,7 +97,7 @@ private:
    }
    void AppendAudioOutBuffer(Kernel::HLERequestContext& ctx) {
-        LOG_WARNING(Service_Audio, "(STUBBED) called");
+        NGLOG_WARNING(Service_Audio, "(STUBBED) called");
        IPC::RequestParser rp{ctx};
        const u64 key{rp.Pop<u64>()};
@@ -108,7 +108,7 @@ private:
    }
    void GetReleasedAudioOutBuffer(Kernel::HLERequestContext& ctx) {
-        LOG_WARNING(Service_Audio, "(STUBBED) called");
+        NGLOG_WARNING(Service_Audio, "(STUBBED) called");
        // TODO(st4rk): This is how libtransistor currently implements the
        // GetReleasedAudioOutBuffer, it should return the key (a VAddr) to the app and this address
@@ -164,7 +164,7 @@ private:
 };
 void AudOutU::ListAudioOuts(Kernel::HLERequestContext& ctx) {
-    LOG_WARNING(Service_Audio, "(STUBBED) called");
+    NGLOG_WARNING(Service_Audio, "(STUBBED) called");
    IPC::RequestParser rp{ctx};
    const std::string audio_interface = "AudioInterface";
@@ -180,7 +180,7 @@ void AudOutU::ListAudioOuts(Kernel::HLERequestContext& ctx) {
 }
 void AudOutU::OpenAudioOut(Kernel::HLERequestContext& ctx) {
-    LOG_WARNING(Service_Audio, "(STUBBED) called");
+    NGLOG_WARNING(Service_Audio, "(STUBBED) called");
    if (!audio_out_interface) {
        audio_out_interface = std::make_shared<IAudioOut>();
diff --git a/src/core/hle/service/audio/audren_u.cpp b/src/core/hle/service/audio/audren_u.cpp
index a9360614c..291885db8 100644
--- a/src/core/hle/service/audio/audren_u.cpp
+++ b/src/core/hle/service/audio/audren_u.cpp
@@ -56,7 +56,7 @@ private:
    }
    void RequestUpdateAudioRenderer(Kernel::HLERequestContext& ctx) {
-        LOG_DEBUG(Service_Audio, "%s", ctx.Description().c_str());
+        NGLOG_DEBUG(Service_Audio, "{}", ctx.Description());
        AudioRendererResponseData response_data{};
        response_data.section_0_size =
@@ -79,7 +79,7 @@ private:
        rb.Push(RESULT_SUCCESS);
-        LOG_WARNING(Service_Audio, "(STUBBED) called");
+        NGLOG_WARNING(Service_Audio, "(STUBBED) called");
    }
    void StartAudioRenderer(Kernel::HLERequestContext& ctx) {
@@ -87,7 +87,7 @@ private:
        rb.Push(RESULT_SUCCESS);
-        LOG_WARNING(Service_Audio, "(STUBBED) called");
+        NGLOG_WARNING(Service_Audio, "(STUBBED) called");
    }
    void StopAudioRenderer(Kernel::HLERequestContext& ctx) {
@@ -95,7 +95,7 @@ private:
        rb.Push(RESULT_SUCCESS);
-        LOG_WARNING(Service_Audio, "(STUBBED) called");
+        NGLOG_WARNING(Service_Audio, "(STUBBED) called");
    }
    void QuerySystemEvent(Kernel::HLERequestContext& ctx) {
@@ -105,7 +105,7 @@ private:
        rb.Push(RESULT_SUCCESS);
        rb.PushCopyObjects(system_event);
-        LOG_WARNING(Service_Audio, "(STUBBED) called");
+        NGLOG_WARNING(Service_Audio, "(STUBBED) called");
    }
    struct AudioRendererStateEntry {
@@ -177,7 +177,7 @@ public:
 private:
    void ListAudioDeviceName(Kernel::HLERequestContext& ctx) {
-        LOG_WARNING(Service_Audio, "(STUBBED) called");
+        NGLOG_WARNING(Service_Audio, "(STUBBED) called");
        IPC::RequestParser rp{ctx};
        const std::string audio_interface = "AudioInterface";
@@ -189,7 +189,7 @@ private:
    }
    void SetAudioDeviceOutputVolume(Kernel::HLERequestContext& ctx) {
-        LOG_WARNING(Service_Audio, "(STUBBED) called");
+        NGLOG_WARNING(Service_Audio, "(STUBBED) called");
        IPC::RequestParser rp{ctx};
        f32 volume = static_cast<f32>(rp.Pop<u32>());
@@ -202,7 +202,7 @@ private:
    }
    void GetActiveAudioDeviceName(Kernel::HLERequestContext& ctx) {
-        LOG_WARNING(Service_Audio, "(STUBBED) called");
+        NGLOG_WARNING(Service_Audio, "(STUBBED) called");
        IPC::RequestParser rp{ctx};
        const std::string audio_interface = "AudioDevice";
@@ -214,7 +214,7 @@ private:
    }
    void QueryAudioDeviceSystemEvent(Kernel::HLERequestContext& ctx) {
-        LOG_WARNING(Service_Audio, "(STUBBED) called");
+        NGLOG_WARNING(Service_Audio, "(STUBBED) called");
        buffer_event->Signal();
@@ -224,7 +224,7 @@ private:
    }
    void GetActiveChannelCount(Kernel::HLERequestContext& ctx) {
-        LOG_WARNING(Service_Audio, "(STUBBED) called");
+        NGLOG_WARNING(Service_Audio, "(STUBBED) called");
        IPC::ResponseBuilder rb{ctx, 3};
        rb.Push(RESULT_SUCCESS);
        rb.Push<u32>(1);
@@ -251,7 +251,7 @@ void AudRenU::OpenAudioRenderer(Kernel::HLERequestContext& ctx) {
    rb.Push(RESULT_SUCCESS);
    rb.PushIpcInterface<Audio::IAudioRenderer>();
-    LOG_DEBUG(Service_Audio, "called");
+    NGLOG_DEBUG(Service_Audio, "called");
 }
 void AudRenU::GetAudioRendererWorkBufferSize(Kernel::HLERequestContext& ctx) {
@@ -260,7 +260,7 @@ void AudRenU::GetAudioRendererWorkBufferSize(Kernel::HLERequestContext& ctx) {
    rb.Push(RESULT_SUCCESS);
    rb.Push<u64>(0x4000);
-    LOG_WARNING(Service_Audio, "(STUBBED) called");
+    NGLOG_WARNING(Service_Audio, "(STUBBED) called");
 }
 void AudRenU::GetAudioDevice(Kernel::HLERequestContext& ctx) {
@@ -269,7 +269,7 @@ void AudRenU::GetAudioDevice(Kernel::HLERequestContext& ctx) {
    rb.Push(RESULT_SUCCESS);
    rb.PushIpcInterface<Audio::IAudioDevice>();
-    LOG_DEBUG(Service_Audio, "called");
+    NGLOG_DEBUG(Service_Audio, "called");
 }
 } // namespace Service::Audio
diff --git a/src/core/hle/service/fatal/fatal.cpp b/src/core/hle/service/fatal/fatal.cpp
index 41d58f999..d7ad0600a 100644
--- a/src/core/hle/service/fatal/fatal.cpp
+++ b/src/core/hle/service/fatal/fatal.cpp
@@ -16,13 +16,13 @@ Module::Interface::Interface(std::shared_ptr<Module> module, const char* name)
 void Module::Interface::FatalSimple(Kernel::HLERequestContext& ctx) {
    IPC::RequestParser rp(ctx);
    u32 error_code = rp.Pop<u32>();
-    LOG_WARNING(Service_Fatal, "(STUBBED) called, error_code=0x%X", error_code);
+    NGLOG_WARNING(Service_Fatal, "(STUBBED) called, error_code={:#X}", error_code);
    IPC::ResponseBuilder rb{ctx, 2};
    rb.Push(RESULT_SUCCESS);
 }
 void Module::Interface::TransitionToFatalError(Kernel::HLERequestContext& ctx) {
-    LOG_WARNING(Service_Fatal, "(STUBBED) called");
+    NGLOG_WARNING(Service_Fatal, "(STUBBED) called");
    IPC::ResponseBuilder rb{ctx, 2};
    rb.Push(RESULT_SUCCESS);
 }
diff --git a/src/core/hle/service/filesystem/filesystem.cpp b/src/core/hle/service/filesystem/filesystem.cpp
index 9e504992f..c2951c560 100644
--- a/src/core/hle/service/filesystem/filesystem.cpp
+++ b/src/core/hle/service/filesystem/filesystem.cpp
@@ -25,14 +25,14 @@ ResultCode RegisterFileSystem(std::unique_ptr<FileSys::FileSystemFactory>&& fact
    ASSERT_MSG(inserted, "Tried to register more than one system with same id code");
    auto& filesystem = result.first->second;
-    LOG_DEBUG(Service_FS, "Registered file system %s with id code 0x%08X",
+    NGLOG_DEBUG(Service_FS, "Registered file system {} with id code {:#010X}",
-              filesystem->GetName().c_str(), static_cast<u32>(type));
+                filesystem->GetName(), static_cast<u32>(type));
    return RESULT_SUCCESS;
 }
 ResultVal<std::unique_ptr<FileSys::FileSystemBackend>> OpenFileSystem(Type type,
                                                                      FileSys::Path& path) {
-    LOG_TRACE(Service_FS, "Opening FileSystem with type=%d", type);
+    NGLOG_TRACE(Service_FS, "Opening FileSystem with type={}", static_cast<u32>(type));
    auto itr = filesystem_map.find(type);
    if (itr == filesystem_map.end()) {
@@ -44,7 +44,7 @@ ResultVal<std::unique_ptr<FileSys::FileSystemBackend>> OpenFileSystem(Type type,
 }
 ResultCode FormatFileSystem(Type type) {
-    LOG_TRACE(Service_FS, "Formatting FileSystem with type=%d", type);
+    NGLOG_TRACE(Service_FS, "Formatting FileSystem with type={}", static_cast<u32>(type));
    auto itr = filesystem_map.find(type);
    if (itr == filesystem_map.end()) {
diff --git a/src/core/hle/service/filesystem/fsp_srv.cpp b/src/core/hle/service/filesystem/fsp_srv.cpp
index 2f476c869..ed9f2ef72 100644
--- a/src/core/hle/service/filesystem/fsp_srv.cpp
+++ b/src/core/hle/service/filesystem/fsp_srv.cpp
@@ -35,7 +35,7 @@ private:
        const s64 offset = rp.Pop<s64>();
        const s64 length = rp.Pop<s64>();
-        LOG_DEBUG(Service_FS, "called, offset=0x%ld, length=0x%ld", offset, length);
+        NGLOG_DEBUG(Service_FS, "called, offset={:#X}, length={}", offset, length);
        // Error checking
        if (length < 0) {
@@ -87,7 +87,7 @@ private:
        const s64 offset = rp.Pop<s64>();
        const s64 length = rp.Pop<s64>();
-        LOG_DEBUG(Service_FS, "called, offset=0x%ld, length=0x%ld", offset, length);
+        NGLOG_DEBUG(Service_FS, "called, offset={:#X}, length={}", offset, length);
        // Error checking
        if (length < 0) {
@@ -124,7 +124,7 @@ private:
        const s64 offset = rp.Pop<s64>();
        const s64 length = rp.Pop<s64>();
-        LOG_DEBUG(Service_FS, "called, offset=0x%ld, length=0x%ld", offset, length);
+        NGLOG_DEBUG(Service_FS, "called, offset={:#X}, length={}", offset, length);
        // Error checking
        if (length < 0) {
@@ -152,7 +152,7 @@ private:
    }
    void Flush(Kernel::HLERequestContext& ctx) {
-        LOG_DEBUG(Service_FS, "called");
+        NGLOG_DEBUG(Service_FS, "called");
        backend->Flush();
        IPC::ResponseBuilder rb{ctx, 2};
@@ -163,7 +163,7 @@ private:
        IPC::RequestParser rp{ctx};
        const u64 size = rp.Pop<u64>();
        backend->SetSize(size);
-        LOG_DEBUG(Service_FS, "called, size=%" PRIu64, size);
+        NGLOG_DEBUG(Service_FS, "called, size={}", size);
        IPC::ResponseBuilder rb{ctx, 2};
        rb.Push(RESULT_SUCCESS);
@@ -171,7 +171,7 @@ private:
    void GetSize(Kernel::HLERequestContext& ctx) {
        const u64 size = backend->GetSize();
-        LOG_DEBUG(Service_FS, "called, size=%" PRIu64, size);
+        NGLOG_DEBUG(Service_FS, "called, size={}", size);
        IPC::ResponseBuilder rb{ctx, 4};
        rb.Push(RESULT_SUCCESS);
@@ -197,7 +197,7 @@ private:
        IPC::RequestParser rp{ctx};
        const u64 unk = rp.Pop<u64>();
-        LOG_DEBUG(Service_FS, "called, unk=0x%llx", unk);
+        NGLOG_DEBUG(Service_FS, "called, unk={:#X}", unk);
        // Calculate how many entries we can fit in the output buffer
        u64 count_entries = ctx.GetWriteBufferSize() / sizeof(FileSys::Entry);
@@ -219,7 +219,7 @@ private:
    }
    void GetEntryCount(Kernel::HLERequestContext& ctx) {
-        LOG_DEBUG(Service_FS, "called");
+        NGLOG_DEBUG(Service_FS, "called");
        u64 count = backend->GetEntryCount();
@@ -239,7 +239,7 @@ public:
            {2, &IFileSystem::CreateDirectory, "CreateDirectory"},
            {3, nullptr, "DeleteDirectory"},
            {4, nullptr, "DeleteDirectoryRecursively"},
-            {5, nullptr, "RenameFile"},
+            {5, &IFileSystem::RenameFile, "RenameFile"},
            {6, nullptr, "RenameDirectory"},
            {7, &IFileSystem::GetEntryType, "GetEntryType"},
            {8, &IFileSystem::OpenFile, "OpenFile"},
@@ -265,8 +265,7 @@ public:
        u64 mode = rp.Pop<u64>();
        u32 size = rp.Pop<u32>();
-        LOG_DEBUG(Service_FS, "called file %s mode 0x%" PRIX64 " size 0x%08X", name.c_str(), mode,
+        NGLOG_DEBUG(Service_FS, "called file {} mode {:#X} size {:#010X}", name, mode, size);
-                  size);
        IPC::ResponseBuilder rb{ctx, 2};
        rb.Push(backend->CreateFile(name, size));
@@ -280,7 +279,7 @@ public:
        std::string name(file_buffer.begin(), end);
-        LOG_DEBUG(Service_FS, "called file %s", name.c_str());
+        NGLOG_DEBUG(Service_FS, "called file {}", name);
        IPC::ResponseBuilder rb{ctx, 2};
        rb.Push(backend->DeleteFile(name));
@@ -294,12 +293,32 @@ public:
        std::string name(file_buffer.begin(), end);
-        LOG_DEBUG(Service_FS, "called directory %s", name.c_str());
+        NGLOG_DEBUG(Service_FS, "called directory {}", name);
        IPC::ResponseBuilder rb{ctx, 2};
        rb.Push(backend->CreateDirectory(name));
    }
+    void RenameFile(Kernel::HLERequestContext& ctx) {
+        IPC::RequestParser rp{ctx};
+        std::vector<u8> buffer;
+        buffer.resize(ctx.BufferDescriptorX()[0].Size());
+        Memory::ReadBlock(ctx.BufferDescriptorX()[0].Address(), buffer.data(), buffer.size());
+        auto end = std::find(buffer.begin(), buffer.end(), '\0');
+        std::string src_name(buffer.begin(), end);
+        buffer.resize(ctx.BufferDescriptorX()[1].Size());
+        Memory::ReadBlock(ctx.BufferDescriptorX()[1].Address(), buffer.data(), buffer.size());
+        end = std::find(buffer.begin(), buffer.end(), '\0');
+        std::string dst_name(buffer.begin(), end);
+        NGLOG_DEBUG(Service_FS, "called file '{}' to file '{}'", src_name, dst_name);
+        IPC::ResponseBuilder rb{ctx, 2};
+        rb.Push(backend->RenameFile(src_name, dst_name));
+    }
    void OpenFile(Kernel::HLERequestContext& ctx) {
        IPC::RequestParser rp{ctx};
@@ -310,7 +329,7 @@ public:
        auto mode = static_cast<FileSys::Mode>(rp.Pop<u32>());
-        LOG_DEBUG(Service_FS, "called file %s mode %u", name.c_str(), static_cast<u32>(mode));
+        NGLOG_DEBUG(Service_FS, "called file {} mode {}", name, static_cast<u32>(mode));
        auto result = backend->OpenFile(name, mode);
        if (result.Failed()) {
@@ -337,7 +356,7 @@ public:
        // TODO(Subv): Implement this filter.
        u32 filter_flags = rp.Pop<u32>();
-        LOG_DEBUG(Service_FS, "called directory %s filter %u", name.c_str(), filter_flags);
+        NGLOG_DEBUG(Service_FS, "called directory {} filter {}", name, filter_flags);
        auto result = backend->OpenDirectory(name);
        if (result.Failed()) {
@@ -361,7 +380,7 @@ public:
        std::string name(file_buffer.begin(), end);
-        LOG_DEBUG(Service_FS, "called file %s", name.c_str());
+        NGLOG_DEBUG(Service_FS, "called file {}", name);
        auto result = backend->GetEntryType(name);
        if (result.Failed()) {
@@ -376,7 +395,7 @@ public:
    }
    void Commit(Kernel::HLERequestContext& ctx) {
-        LOG_WARNING(Service_FS, "(STUBBED) called");
+        NGLOG_WARNING(Service_FS, "(STUBBED) called");
        IPC::ResponseBuilder rb{ctx, 2};
        rb.Push(RESULT_SUCCESS);
@@ -492,14 +511,14 @@ void FSP_SRV::TryLoadRomFS() {
 }
 void FSP_SRV::Initialize(Kernel::HLERequestContext& ctx) {
-    LOG_WARNING(Service_FS, "(STUBBED) called");
+    NGLOG_WARNING(Service_FS, "(STUBBED) called");
    IPC::ResponseBuilder rb{ctx, 2};
    rb.Push(RESULT_SUCCESS);
 }
 void FSP_SRV::MountSdCard(Kernel::HLERequestContext& ctx) {
-    LOG_DEBUG(Service_FS, "called");
+    NGLOG_DEBUG(Service_FS, "called");
    FileSys::Path unused;
    auto filesystem = OpenFileSystem(Type::SDMC, unused).Unwrap();
@@ -516,14 +535,14 @@ void FSP_SRV::CreateSaveData(Kernel::HLERequestContext& ctx) {
    auto save_create_struct = rp.PopRaw<std::array<u8, 0x40>>();
    u128 uid = rp.PopRaw<u128>();
-    LOG_WARNING(Service_FS, "(STUBBED) called uid = %016" PRIX64 "%016" PRIX64, uid[1], uid[0]);
+    NGLOG_WARNING(Service_FS, "(STUBBED) called uid = {:016X}{:016X}", uid[1], uid[0]);
    IPC::ResponseBuilder rb{ctx, 2};
    rb.Push(RESULT_SUCCESS);
 }
 void FSP_SRV::MountSaveData(Kernel::HLERequestContext& ctx) {
-    LOG_WARNING(Service_FS, "(STUBBED) called");
+    NGLOG_WARNING(Service_FS, "(STUBBED) called");
    FileSys::Path unused;
    auto filesystem = OpenFileSystem(Type::SaveData, unused).Unwrap();
@@ -534,7 +553,7 @@ void FSP_SRV::MountSaveData(Kernel::HLERequestContext& ctx) {
 }
 void FSP_SRV::GetGlobalAccessLogMode(Kernel::HLERequestContext& ctx) {
-    LOG_WARNING(Service_FS, "(STUBBED) called");
+    NGLOG_WARNING(Service_FS, "(STUBBED) called");
    IPC::ResponseBuilder rb{ctx, 3};
    rb.Push(RESULT_SUCCESS);
@@ -542,12 +561,12 @@ void FSP_SRV::GetGlobalAccessLogMode(Kernel::HLERequestContext& ctx) {
 }
 void FSP_SRV::OpenDataStorageByCurrentProcess(Kernel::HLERequestContext& ctx) {
-    LOG_DEBUG(Service_FS, "called");
+    NGLOG_DEBUG(Service_FS, "called");
    TryLoadRomFS();
    if (!romfs) {
        // TODO (bunnei): Find the right error code to use here
-        LOG_CRITICAL(Service_FS, "no file system interface available!");
+        NGLOG_CRITICAL(Service_FS, "no file system interface available!");
        IPC::ResponseBuilder rb{ctx, 2};
        rb.Push(ResultCode(-1));
        return;
@@ -556,7 +575,7 @@ void FSP_SRV::OpenDataStorageByCurrentProcess(Kernel::HLERequestContext& ctx) {
    // Attempt to open a StorageBackend interface to the RomFS
    auto storage = romfs->OpenFile({}, {});
    if (storage.Failed()) {
-        LOG_CRITICAL(Service_FS, "no storage interface available!");
+        NGLOG_CRITICAL(Service_FS, "no storage interface available!");
        IPC::ResponseBuilder rb{ctx, 2};
        rb.Push(storage.Code());
        return;
@@ -568,7 +587,7 @@ void FSP_SRV::OpenDataStorageByCurrentProcess(Kernel::HLERequestContext& ctx) {
 }
 void FSP_SRV::OpenRomStorage(Kernel::HLERequestContext& ctx) {
-    LOG_WARNING(Service_FS, "(STUBBED) called, using OpenDataStorageByCurrentProcess");
+    NGLOG_WARNING(Service_FS, "(STUBBED) called, using OpenDataStorageByCurrentProcess");
    OpenDataStorageByCurrentProcess(ctx);
 }
diff --git a/src/core/hle/service/friend/friend.cpp b/src/core/hle/service/friend/friend.cpp
index c98a46e05..94d9fbf25 100644
--- a/src/core/hle/service/friend/friend.cpp
+++ b/src/core/hle/service/friend/friend.cpp
@@ -13,7 +13,7 @@ namespace Service::Friend {
 void Module::Interface::CreateFriendService(Kernel::HLERequestContext& ctx) {
    IPC::ResponseBuilder rb{ctx, 2};
    rb.Push(RESULT_SUCCESS);
-    LOG_WARNING(Service_Friend, "(STUBBED) called");
+    NGLOG_WARNING(Service_Friend, "(STUBBED) called");
 }
 Module::Interface::Interface(std::shared_ptr<Module> module, const char* name)
diff --git a/src/core/hle/service/hid/hid.cpp b/src/core/hle/service/hid/hid.cpp
index aad5e688b..736180b63 100644
--- a/src/core/hle/service/hid/hid.cpp
+++ b/src/core/hle/service/hid/hid.cpp
@@ -53,7 +53,7 @@ private:
        IPC::ResponseBuilder rb{ctx, 2, 1};
        rb.Push(RESULT_SUCCESS);
        rb.PushCopyObjects(shared_mem);
-        LOG_DEBUG(Service_HID, "called");
+        NGLOG_DEBUG(Service_HID, "called");
    }
    void LoadInputDevices() {
@@ -184,7 +184,7 @@ private:
    void ActivateVibrationDevice(Kernel::HLERequestContext& ctx) {
        IPC::ResponseBuilder rb{ctx, 2};
        rb.Push(RESULT_SUCCESS);
-        LOG_WARNING(Service_HID, "(STUBBED) called");
+        NGLOG_WARNING(Service_HID, "(STUBBED) called");
    }
 };
@@ -286,144 +286,144 @@ private:
        IPC::ResponseBuilder rb{ctx, 2, 0, 1};
        rb.Push(RESULT_SUCCESS);
        rb.PushIpcInterface<IAppletResource>(applet_resource);
-        LOG_DEBUG(Service_HID, "called");
+        NGLOG_DEBUG(Service_HID, "called");
    }
    void ActivateDebugPad(Kernel::HLERequestContext& ctx) {
        IPC::ResponseBuilder rb{ctx, 2};
        rb.Push(RESULT_SUCCESS);
-        LOG_WARNING(Service_HID, "(STUBBED) called");
+        NGLOG_WARNING(Service_HID, "(STUBBED) called");
    }
    void ActivateTouchScreen(Kernel::HLERequestContext& ctx) {
        IPC::ResponseBuilder rb{ctx, 2};
        rb.Push(RESULT_SUCCESS);
-        LOG_WARNING(Service_HID, "(STUBBED) called");
+        NGLOG_WARNING(Service_HID, "(STUBBED) called");
    }
    void ActivateMouse(Kernel::HLERequestContext& ctx) {
        IPC::ResponseBuilder rb{ctx, 2};
        rb.Push(RESULT_SUCCESS);
-        LOG_WARNING(Service_HID, "(STUBBED) called");
+        NGLOG_WARNING(Service_HID, "(STUBBED) called");
    }
    void ActivateKeyboard(Kernel::HLERequestContext& ctx) {
        IPC::ResponseBuilder rb{ctx, 2};
        rb.Push(RESULT_SUCCESS);
-        LOG_WARNING(Service_HID, "(STUBBED) called");
+        NGLOG_WARNING(Service_HID, "(STUBBED) called");
    }
    void StartSixAxisSensor(Kernel::HLERequestContext& ctx) {
        IPC::ResponseBuilder rb{ctx, 2};
        rb.Push(RESULT_SUCCESS);
-        LOG_WARNING(Service_HID, "(STUBBED) called");
+        NGLOG_WARNING(Service_HID, "(STUBBED) called");
    }
    void SetGyroscopeZeroDriftMode(Kernel::HLERequestContext& ctx) {
        IPC::ResponseBuilder rb{ctx, 2};
        rb.Push(RESULT_SUCCESS);
-        LOG_WARNING(Service_HID, "(STUBBED) called");
+        NGLOG_WARNING(Service_HID, "(STUBBED) called");
    }
    void SetSupportedNpadStyleSet(Kernel::HLERequestContext& ctx) {
        IPC::ResponseBuilder rb{ctx, 2};
        rb.Push(RESULT_SUCCESS);
-        LOG_WARNING(Service_HID, "(STUBBED) called");
+        NGLOG_WARNING(Service_HID, "(STUBBED) called");
    }
    void GetSupportedNpadStyleSet(Kernel::HLERequestContext& ctx) {
        IPC::ResponseBuilder rb{ctx, 3};
        rb.Push(RESULT_SUCCESS);
        rb.Push<u32>(0);
-        LOG_WARNING(Service_HID, "(STUBBED) called");
+        NGLOG_WARNING(Service_HID, "(STUBBED) called");
    }
    void SetSupportedNpadIdType(Kernel::HLERequestContext& ctx) {
        IPC::ResponseBuilder rb{ctx, 2};
        rb.Push(RESULT_SUCCESS);
-        LOG_WARNING(Service_HID, "(STUBBED) called");
+        NGLOG_WARNING(Service_HID, "(STUBBED) called");
    }
    void ActivateNpad(Kernel::HLERequestContext& ctx) {
        IPC::ResponseBuilder rb{ctx, 2};
        rb.Push(RESULT_SUCCESS);
-        LOG_WARNING(Service_HID, "(STUBBED) called");
+        NGLOG_WARNING(Service_HID, "(STUBBED) called");
    }
    void AcquireNpadStyleSetUpdateEventHandle(Kernel::HLERequestContext& ctx) {
        IPC::ResponseBuilder rb{ctx, 2, 1};
        rb.Push(RESULT_SUCCESS);
        rb.PushCopyObjects(event);
-        LOG_WARNING(Service_HID, "(STUBBED) called");
+        NGLOG_WARNING(Service_HID, "(STUBBED) called");
    }
    void GetPlayerLedPattern(Kernel::HLERequestContext& ctx) {
        IPC::ResponseBuilder rb{ctx, 2};
        rb.Push(RESULT_SUCCESS);
-        LOG_WARNING(Service_HID, "(STUBBED) called");
+        NGLOG_WARNING(Service_HID, "(STUBBED) called");
    }
    void SetNpadJoyHoldType(Kernel::HLERequestContext& ctx) {
        IPC::ResponseBuilder rb{ctx, 2};
        rb.Push(RESULT_SUCCESS);
-        LOG_WARNING(Service_HID, "(STUBBED) called");
+        NGLOG_WARNING(Service_HID, "(STUBBED) called");
    }
    void GetNpadJoyHoldType(Kernel::HLERequestContext& ctx) {
        IPC::ResponseBuilder rb{ctx, 3};
        rb.Push(RESULT_SUCCESS);
        rb.Push(joy_hold_type);
-        LOG_WARNING(Service_HID, "(STUBBED) called");
+        NGLOG_WARNING(Service_HID, "(STUBBED) called");
    }
    void SetNpadJoyAssignmentModeSingleByDefault(Kernel::HLERequestContext& ctx) {
        IPC::ResponseBuilder rb{ctx, 2};
        rb.Push(RESULT_SUCCESS);
-        LOG_WARNING(Service_HID, "(STUBBED) called");
+        NGLOG_WARNING(Service_HID, "(STUBBED) called");
    }
    void SendVibrationValue(Kernel::HLERequestContext& ctx) {
        IPC::ResponseBuilder rb{ctx, 2};
        rb.Push(RESULT_SUCCESS);
-        LOG_WARNING(Service_HID, "(STUBBED) called");
+        NGLOG_WARNING(Service_HID, "(STUBBED) called");
    }
    void GetActualVibrationValue(Kernel::HLERequestContext& ctx) {
        IPC::ResponseBuilder rb{ctx, 2};
        rb.Push(RESULT_SUCCESS);
-        LOG_WARNING(Service_HID, "(STUBBED) called");
+        NGLOG_WARNING(Service_HID, "(STUBBED) called");
    }
    void SetNpadJoyAssignmentModeDual(Kernel::HLERequestContext& ctx) {
        IPC::ResponseBuilder rb{ctx, 2};
        rb.Push(RESULT_SUCCESS);
-        LOG_WARNING(Service_HID, "(STUBBED) called");
+        NGLOG_WARNING(Service_HID, "(STUBBED) called");
    }
    void SetNpadHandheldActivationMode(Kernel::HLERequestContext& ctx) {
        IPC::ResponseBuilder rb{ctx, 2};
        rb.Push(RESULT_SUCCESS);
-        LOG_WARNING(Service_HID, "(STUBBED) called");
+        NGLOG_WARNING(Service_HID, "(STUBBED) called");
    }
    void GetVibrationDeviceInfo(Kernel::HLERequestContext& ctx) {
        IPC::ResponseBuilder rb{ctx, 4};
        rb.Push(RESULT_SUCCESS);
        rb.Push<u64>(0);
-        LOG_WARNING(Service_HID, "(STUBBED) called");
+        NGLOG_WARNING(Service_HID, "(STUBBED) called");
    }
    void CreateActiveVibrationDeviceList(Kernel::HLERequestContext& ctx) {
        IPC::ResponseBuilder rb{ctx, 2, 0, 1};
        rb.Push(RESULT_SUCCESS);
        rb.PushIpcInterface<IActiveVibrationDeviceList>();
-        LOG_DEBUG(Service_HID, "called");
+        NGLOG_DEBUG(Service_HID, "called");
    }
    void SendVibrationValues(Kernel::HLERequestContext& ctx) {
        IPC::ResponseBuilder rb{ctx, 2};
        rb.Push(RESULT_SUCCESS);
-        LOG_WARNING(Service_HID, "(STUBBED) called");
+        NGLOG_WARNING(Service_HID, "(STUBBED) called");
    }
 };
diff --git a/src/core/hle/service/lm/lm.cpp b/src/core/hle/service/lm/lm.cpp
index b87172dff..46194643e 100644
--- a/src/core/hle/service/lm/lm.cpp
+++ b/src/core/hle/service/lm/lm.cpp
@@ -141,19 +141,19 @@ private:
        if (header.IsTailLog()) {
            switch (header.severity) {
            case MessageHeader::Severity::Trace:
-                LOG_TRACE(Debug_Emulated, "%s", log_stream.str().c_str());
+                NGLOG_TRACE(Debug_Emulated, "{}", log_stream.str());
                break;
            case MessageHeader::Severity::Info:
-                LOG_INFO(Debug_Emulated, "%s", log_stream.str().c_str());
+                NGLOG_INFO(Debug_Emulated, "{}", log_stream.str());
                break;
            case MessageHeader::Severity::Warning:
-                LOG_WARNING(Debug_Emulated, "%s", log_stream.str().c_str());
+                NGLOG_WARNING(Debug_Emulated, "{}", log_stream.str());
                break;
            case MessageHeader::Severity::Error:
-                LOG_ERROR(Debug_Emulated, "%s", log_stream.str().c_str());
+                NGLOG_ERROR(Debug_Emulated, "{}", log_stream.str());
                break;
            case MessageHeader::Severity::Critical:
-                LOG_CRITICAL(Debug_Emulated, "%s", log_stream.str().c_str());
+                NGLOG_CRITICAL(Debug_Emulated, "{}", log_stream.str());
                break;
            }
        }
@@ -178,7 +178,7 @@ void LM::Initialize(Kernel::HLERequestContext& ctx) {
    rb.Push(RESULT_SUCCESS);
    rb.PushIpcInterface<Logger>();
-    LOG_DEBUG(Service_LM, "called");
+    NGLOG_DEBUG(Service_LM, "called");
 }
 LM::LM() : ServiceFramework("lm") {
diff --git a/src/core/hle/service/nifm/nifm.cpp b/src/core/hle/service/nifm/nifm.cpp
index df1e7f8fe..eee92cfcd 100644
--- a/src/core/hle/service/nifm/nifm.cpp
+++ b/src/core/hle/service/nifm/nifm.cpp
@@ -62,24 +62,24 @@ public:
 private:
    void GetRequestState(Kernel::HLERequestContext& ctx) {
-        LOG_WARNING(Service_NIFM, "(STUBBED) called");
+        NGLOG_WARNING(Service_NIFM, "(STUBBED) called");
        IPC::ResponseBuilder rb{ctx, 3};
        rb.Push(RESULT_SUCCESS);
        rb.Push<u32>(0);
    }
    void GetResult(Kernel::HLERequestContext& ctx) {
-        LOG_WARNING(Service_NIFM, "(STUBBED) called");
+        NGLOG_WARNING(Service_NIFM, "(STUBBED) called");
        IPC::ResponseBuilder rb{ctx, 2};
        rb.Push(RESULT_SUCCESS);
    }
    void GetSystemEventReadableHandles(Kernel::HLERequestContext& ctx) {
-        LOG_WARNING(Service_NIFM, "(STUBBED) called");
+        NGLOG_WARNING(Service_NIFM, "(STUBBED) called");
        IPC::ResponseBuilder rb{ctx, 2, 2};
        rb.Push(RESULT_SUCCESS);
        rb.PushCopyObjects(event1, event2);
    }
    void Cancel(Kernel::HLERequestContext& ctx) {
-        LOG_WARNING(Service_NIFM, "(STUBBED) called");
+        NGLOG_WARNING(Service_NIFM, "(STUBBED) called");
        IPC::ResponseBuilder rb{ctx, 2};
        rb.Push(RESULT_SUCCESS);
    }
@@ -105,7 +105,7 @@ public:
 private:
    void GetClientId(Kernel::HLERequestContext& ctx) {
-        LOG_WARNING(Service_NIFM, "(STUBBED) called");
+        NGLOG_WARNING(Service_NIFM, "(STUBBED) called");
        IPC::ResponseBuilder rb{ctx, 4};
        rb.Push(RESULT_SUCCESS);
        rb.Push<u64>(0);
@@ -116,7 +116,7 @@ private:
        rb.Push(RESULT_SUCCESS);
        rb.PushIpcInterface<IScanRequest>();
-        LOG_DEBUG(Service_NIFM, "called");
+        NGLOG_DEBUG(Service_NIFM, "called");
    }
    void CreateRequest(Kernel::HLERequestContext& ctx) {
        IPC::ResponseBuilder rb{ctx, 2, 0, 1};
@@ -124,10 +124,10 @@ private:
        rb.Push(RESULT_SUCCESS);
        rb.PushIpcInterface<IRequest>();
-        LOG_DEBUG(Service_NIFM, "called");
+        NGLOG_DEBUG(Service_NIFM, "called");
    }
    void RemoveNetworkProfile(Kernel::HLERequestContext& ctx) {
-        LOG_WARNING(Service_NIFM, "(STUBBED) called");
+        NGLOG_WARNING(Service_NIFM, "(STUBBED) called");
        IPC::ResponseBuilder rb{ctx, 2};
        rb.Push(RESULT_SUCCESS);
    }
@@ -137,7 +137,7 @@ private:
        rb.Push(RESULT_SUCCESS);
        rb.PushIpcInterface<INetworkProfile>();
-        LOG_DEBUG(Service_NIFM, "called");
+        NGLOG_DEBUG(Service_NIFM, "called");
    }
 };
@@ -187,14 +187,14 @@ void Module::Interface::CreateGeneralServiceOld(Kernel::HLERequestContext& ctx)
    IPC::ResponseBuilder rb{ctx, 2, 0, 1};
    rb.Push(RESULT_SUCCESS);
    rb.PushIpcInterface<IGeneralService>();
-    LOG_DEBUG(Service_NIFM, "called");
+    NGLOG_DEBUG(Service_NIFM, "called");
 }
 void Module::Interface::CreateGeneralService(Kernel::HLERequestContext& ctx) {
    IPC::ResponseBuilder rb{ctx, 2, 0, 1};
    rb.Push(RESULT_SUCCESS);
    rb.PushIpcInterface<IGeneralService>();
-    LOG_DEBUG(Service_NIFM, "called");
+    NGLOG_DEBUG(Service_NIFM, "called");
 }
 Module::Interface::Interface(std::shared_ptr<Module> module, const char* name)
diff --git a/src/core/hle/service/ns/pl_u.cpp b/src/core/hle/service/ns/pl_u.cpp
index c416ad720..c2a647e89 100644
--- a/src/core/hle/service/ns/pl_u.cpp
+++ b/src/core/hle/service/ns/pl_u.cpp
@@ -52,7 +52,7 @@ PL_U::PL_U() : ServiceFramework("pl:u") {
        ASSERT(file.GetSize() == SHARED_FONT_MEM_SIZE);
        file.ReadBytes(shared_font->data(), shared_font->size());
    } else {
-        LOG_WARNING(Service_NS, "Unable to load shared font: %s", filepath.c_str());
+        NGLOG_WARNING(Service_NS, "Unable to load shared font: {}", filepath);
    }
 }
@@ -60,7 +60,7 @@ void PL_U::RequestLoad(Kernel::HLERequestContext& ctx) {
    IPC::RequestParser rp{ctx};
    const u32 shared_font_type{rp.Pop<u32>()};
-    LOG_DEBUG(Service_NS, "called, shared_font_type=%d", shared_font_type);
+    NGLOG_DEBUG(Service_NS, "called, shared_font_type={}", shared_font_type);
    IPC::ResponseBuilder rb{ctx, 2};
    rb.Push(RESULT_SUCCESS);
 }
@@ -69,7 +69,7 @@ void PL_U::GetLoadState(Kernel::HLERequestContext& ctx) {
    IPC::RequestParser rp{ctx};
    const u32 font_id{rp.Pop<u32>()};
-    LOG_DEBUG(Service_NS, "called, font_id=%d", font_id);
+    NGLOG_DEBUG(Service_NS, "called, font_id={}", font_id);
    IPC::ResponseBuilder rb{ctx, 3};
    rb.Push(RESULT_SUCCESS);
    rb.Push<u32>(static_cast<u32>(LoadState::Done));
@@ -79,7 +79,7 @@ void PL_U::GetSize(Kernel::HLERequestContext& ctx) {
    IPC::RequestParser rp{ctx};
    const u32 font_id{rp.Pop<u32>()};
-    LOG_DEBUG(Service_NS, "called, font_id=%d", font_id);
+    NGLOG_DEBUG(Service_NS, "called, font_id={}", font_id);
    IPC::ResponseBuilder rb{ctx, 3};
    rb.Push(RESULT_SUCCESS);
    rb.Push<u32>(SHARED_FONT_REGIONS[font_id].size);
@@ -89,7 +89,7 @@ void PL_U::GetSharedMemoryAddressOffset(Kernel::HLERequestContext& ctx) {
    IPC::RequestParser rp{ctx};
    const u32 font_id{rp.Pop<u32>()};
-    LOG_DEBUG(Service_NS, "called, font_id=%d", font_id);
+    NGLOG_DEBUG(Service_NS, "called, font_id={}", font_id);
    IPC::ResponseBuilder rb{ctx, 3};
    rb.Push(RESULT_SUCCESS);
    rb.Push<u32>(SHARED_FONT_REGIONS[font_id].offset);
@@ -110,7 +110,7 @@ void PL_U::GetSharedMemoryNativeHandle(Kernel::HLERequestContext& ctx) {
        Kernel::MemoryPermission::Read, SHARED_FONT_MEM_VADDR, Kernel::MemoryRegion::BASE,
        "PL_U:shared_font_mem");
-    LOG_DEBUG(Service_NS, "called");
+    NGLOG_DEBUG(Service_NS, "called");
    IPC::ResponseBuilder rb{ctx, 2, 1};
    rb.Push(RESULT_SUCCESS);
    rb.PushCopyObjects(shared_font_mem);
diff --git a/src/core/hle/service/nvdrv/devices/nvdisp_disp0.cpp b/src/core/hle/service/nvdrv/devices/nvdisp_disp0.cpp
index 61f22b1a5..103e66d0c 100644
--- a/src/core/hle/service/nvdrv/devices/nvdisp_disp0.cpp
+++ b/src/core/hle/service/nvdrv/devices/nvdisp_disp0.cpp
@@ -13,16 +13,16 @@
 namespace Service::Nvidia::Devices {
 u32 nvdisp_disp0::ioctl(Ioctl command, const std::vector<u8>& input, std::vector<u8>& output) {
-    UNIMPLEMENTED();
+    UNIMPLEMENTED_MSG("Unimplemented ioctl");
    return 0;
 }
 void nvdisp_disp0::flip(u32 buffer_handle, u32 offset, u32 format, u32 width, u32 height,
                        u32 stride, NVFlinger::BufferQueue::BufferTransformFlags transform) {
    VAddr addr = nvmap_dev->GetObjectAddress(buffer_handle);
-    LOG_WARNING(Service,
+    NGLOG_WARNING(Service,
-                "Drawing from address %lx offset %08X Width %u Height %u Stride %u Format %u", addr,
+                  "Drawing from address {:X} offset {:08X} Width {} Height {} Stride {} Format {}",
-                offset, width, height, stride, format);
+                  addr, offset, width, height, stride, format);
    using PixelFormat = Tegra::FramebufferConfig::PixelFormat;
    const Tegra::FramebufferConfig framebuffer{
diff --git a/src/core/hle/service/nvdrv/devices/nvhost_as_gpu.cpp b/src/core/hle/service/nvdrv/devices/nvhost_as_gpu.cpp
index 71e844959..36d7f837b 100644
--- a/src/core/hle/service/nvdrv/devices/nvhost_as_gpu.cpp
+++ b/src/core/hle/service/nvdrv/devices/nvhost_as_gpu.cpp
@@ -12,8 +12,8 @@
 namespace Service::Nvidia::Devices {
 u32 nvhost_as_gpu::ioctl(Ioctl command, const std::vector<u8>& input, std::vector<u8>& output) {
-    LOG_DEBUG(Service_NVDRV, "called, command=0x%08x, input_size=0x%zx, output_size=0x%zx",
+    NGLOG_DEBUG(Service_NVDRV, "called, command={:#010X}, input_size={:#X}, output_size={:#X}",
-              command.raw, input.size(), output.size());
+                command.raw, input.size(), output.size());
    switch (static_cast<IoctlCommand>(command.raw)) {
    case IoctlCommand::IocInitalizeExCommand:
@@ -27,13 +27,18 @@ u32 nvhost_as_gpu::ioctl(Ioctl command, const std::vector<u8>& input, std::vecto
    case IoctlCommand::IocGetVaRegionsCommand:
        return GetVARegions(input, output);
    }
+    if (static_cast<IoctlCommand>(command.cmd.Value()) == IoctlCommand::IocRemapCommand)
+        return Remap(input, output);
+    UNIMPLEMENTED_MSG("Unimplemented ioctl command");
    return 0;
 }
 u32 nvhost_as_gpu::InitalizeEx(const std::vector<u8>& input, std::vector<u8>& output) {
    IoctlInitalizeEx params{};
    std::memcpy(&params, input.data(), input.size());
-    LOG_WARNING(Service_NVDRV, "(STUBBED) called, big_page_size=0x%x", params.big_page_size);
+    NGLOG_WARNING(Service_NVDRV, "(STUBBED) called, big_page_size={:#X}", params.big_page_size);
    std::memcpy(output.data(), &params, output.size());
    return 0;
 }
@@ -41,8 +46,8 @@ u32 nvhost_as_gpu::InitalizeEx(const std::vector<u8>& input, std::vector<u8>& ou
 u32 nvhost_as_gpu::AllocateSpace(const std::vector<u8>& input, std::vector<u8>& output) {
    IoctlAllocSpace params{};
    std::memcpy(&params, input.data(), input.size());
-    LOG_DEBUG(Service_NVDRV, "called, pages=%x, page_size=%x, flags=%x", params.pages,
+    NGLOG_DEBUG(Service_NVDRV, "called, pages={:X}, page_size={:X}, flags={:X}", params.pages,
-              params.page_size, params.flags);
+                params.page_size, params.flags);
    auto& gpu = Core::System::GetInstance().GPU();
    const u64 size{static_cast<u64>(params.pages) * static_cast<u64>(params.page_size)};
@@ -56,15 +61,45 @@ u32 nvhost_as_gpu::AllocateSpace(const std::vector<u8>& input, std::vector<u8>&
    return 0;
 }
+u32 nvhost_as_gpu::Remap(const std::vector<u8>& input, std::vector<u8>& output) {
+    size_t num_entries = input.size() / sizeof(IoctlRemapEntry);
+    NGLOG_WARNING(Service_NVDRV, "(STUBBED) called, num_entries=0x{:X}", num_entries);
+    std::vector<IoctlRemapEntry> entries(num_entries);
+    std::memcpy(entries.data(), input.data(), input.size());
+    auto& gpu = Core::System::GetInstance().GPU();
+    for (const auto& entry : entries) {
+        NGLOG_WARNING(Service_NVDRV, "remap entry, offset=0x{:X} handle=0x{:X} pages=0x{:X}",
+                      entry.offset, entry.nvmap_handle, entry.pages);
+        Tegra::GPUVAddr offset = static_cast<Tegra::GPUVAddr>(entry.offset) << 0x10;
+        auto object = nvmap_dev->GetObject(entry.nvmap_handle);
+        ASSERT(object);
+        ASSERT(object->status == nvmap::Object::Status::Allocated);
+        u64 size = static_cast<u64>(entry.pages) << 0x10;
+        ASSERT(size <= object->size);
+        Tegra::GPUVAddr returned = gpu.memory_manager->MapBufferEx(object->addr, offset, size);
+        ASSERT(returned == offset);
+    }
+    std::memcpy(output.data(), entries.data(), output.size());
+    return 0;
+}
 u32 nvhost_as_gpu::MapBufferEx(const std::vector<u8>& input, std::vector<u8>& output) {
    IoctlMapBufferEx params{};
    std::memcpy(&params, input.data(), input.size());
-    LOG_DEBUG(Service_NVDRV,
+    NGLOG_DEBUG(Service_NVDRV,
-              "called, flags=%x, nvmap_handle=%x, buffer_offset=%" PRIu64 ", mapping_size=%" PRIu64
+                "called, flags={:X}, nvmap_handle={:X}, buffer_offset={}, mapping_size={}"
-              ", offset=%" PRIu64,
+                ", offset={}",
-              params.flags, params.nvmap_handle, params.buffer_offset, params.mapping_size,
+                params.flags, params.nvmap_handle, params.buffer_offset, params.mapping_size,
-              params.offset);
+                params.offset);
    if (!params.nvmap_handle) {
        return 0;
@@ -73,6 +108,16 @@ u32 nvhost_as_gpu::MapBufferEx(const std::vector<u8>& input, std::vector<u8>& ou
    auto object = nvmap_dev->GetObject(params.nvmap_handle);
    ASSERT(object);
+    // We can only map objects that have already been assigned a CPU address.
+    ASSERT(object->status == nvmap::Object::Status::Allocated);
+    ASSERT(params.buffer_offset == 0);
+    // The real nvservices doesn't make a distinction between handles and ids, and
+    // object can only have one handle and it will be the same as its id. Assert that this is the
+    // case to prevent unexpected behavior.
+    ASSERT(object->id == params.nvmap_handle);
    auto& gpu = Core::System::GetInstance().GPU();
    if (params.flags & 1) {
@@ -88,7 +133,7 @@ u32 nvhost_as_gpu::MapBufferEx(const std::vector<u8>& input, std::vector<u8>& ou
 u32 nvhost_as_gpu::BindChannel(const std::vector<u8>& input, std::vector<u8>& output) {
    IoctlBindChannel params{};
    std::memcpy(&params, input.data(), input.size());
-    LOG_DEBUG(Service_NVDRV, "called, fd=%x", params.fd);
+    NGLOG_DEBUG(Service_NVDRV, "called, fd={:X}", params.fd);
    channel = params.fd;
    std::memcpy(output.data(), &params, output.size());
    return 0;
@@ -97,8 +142,8 @@ u32 nvhost_as_gpu::BindChannel(const std::vector<u8>& input, std::vector<u8>& ou
 u32 nvhost_as_gpu::GetVARegions(const std::vector<u8>& input, std::vector<u8>& output) {
    IoctlGetVaRegions params{};
    std::memcpy(&params, input.data(), input.size());
-    LOG_WARNING(Service_NVDRV, "(STUBBED) called, buf_addr=%" PRIu64 ", buf_size=%x",
+    NGLOG_WARNING(Service_NVDRV, "(STUBBED) called, buf_addr={:X}, buf_size={:X}", params.buf_addr,
-                params.buf_addr, params.buf_size);
+                  params.buf_size);
    params.buf_size = 0x30;
    params.regions[0].offset = 0x04000000;
diff --git a/src/core/hle/service/nvdrv/devices/nvhost_as_gpu.h b/src/core/hle/service/nvdrv/devices/nvhost_as_gpu.h
index d86c3ebd9..f2dd0c3b3 100644
--- a/src/core/hle/service/nvdrv/devices/nvhost_as_gpu.h
+++ b/src/core/hle/service/nvdrv/devices/nvhost_as_gpu.h
@@ -26,6 +26,7 @@ private:
    enum class IoctlCommand : u32_le {
        IocInitalizeExCommand = 0x40284109,
        IocAllocateSpaceCommand = 0xC0184102,
+        IocRemapCommand = 0x00000014,
        IocMapBufferExCommand = 0xC0284106,
        IocBindChannelCommand = 0x40044101,
        IocGetVaRegionsCommand = 0xC0404108,
@@ -54,6 +55,16 @@ private:
    };
    static_assert(sizeof(IoctlAllocSpace) == 24, "IoctlInitalizeEx is incorrect size");
+    struct IoctlRemapEntry {
+        u16_le flags;
+        u16_le kind;
+        u32_le nvmap_handle;
+        INSERT_PADDING_WORDS(1);
+        u32_le offset;
+        u32_le pages;
+    };
+    static_assert(sizeof(IoctlRemapEntry) == 20, "IoctlRemapEntry is incorrect size");
    struct IoctlMapBufferEx {
        u32_le flags; // bit0: fixed_offset, bit2: cacheable
        u32_le kind;  // -1 is default
@@ -91,6 +102,7 @@ private:
    u32 InitalizeEx(const std::vector<u8>& input, std::vector<u8>& output);
    u32 AllocateSpace(const std::vector<u8>& input, std::vector<u8>& output);
+    u32 Remap(const std::vector<u8>& input, std::vector<u8>& output);
    u32 MapBufferEx(const std::vector<u8>& input, std::vector<u8>& output);
    u32 BindChannel(const std::vector<u8>& input, std::vector<u8>& output);
    u32 GetVARegions(const std::vector<u8>& input, std::vector<u8>& output);
diff --git a/src/core/hle/service/nvdrv/devices/nvhost_ctrl.cpp b/src/core/hle/service/nvdrv/devices/nvhost_ctrl.cpp
index 660a0f665..46f0b6862 100644
--- a/src/core/hle/service/nvdrv/devices/nvhost_ctrl.cpp
+++ b/src/core/hle/service/nvdrv/devices/nvhost_ctrl.cpp
@@ -9,8 +9,8 @@
 namespace Service::Nvidia::Devices {
 u32 nvhost_ctrl::ioctl(Ioctl command, const std::vector<u8>& input, std::vector<u8>& output) {
-    LOG_DEBUG(Service_NVDRV, "called, command=0x%08x, input_size=0x%zx, output_size=0x%zx",
+    NGLOG_DEBUG(Service_NVDRV, "called, command={:#010X}, input_size={:#X}, output_size={:#X}",
-              command.raw, input.size(), output.size());
+                command.raw, input.size(), output.size());
    switch (static_cast<IoctlCommand>(command.raw)) {
    case IoctlCommand::IocGetConfigCommand:
@@ -18,15 +18,15 @@ u32 nvhost_ctrl::ioctl(Ioctl command, const std::vector<u8>& input, std::vector<
    case IoctlCommand::IocCtrlEventWaitCommand:
        return IocCtrlEventWait(input, output);
    }
-    UNIMPLEMENTED();
+    UNIMPLEMENTED_MSG("Unimplemented ioctl");
    return 0;
 }
 u32 nvhost_ctrl::NvOsGetConfigU32(const std::vector<u8>& input, std::vector<u8>& output) {
    IocGetConfigParams params{};
    std::memcpy(&params, input.data(), sizeof(params));
-    LOG_DEBUG(Service_NVDRV, "called, setting=%s!%s", params.domain_str.data(),
+    NGLOG_DEBUG(Service_NVDRV, "called, setting={}!{}", params.domain_str.data(),
-              params.param_str.data());
+                params.param_str.data());
    if (!strcmp(params.domain_str.data(), "nv")) {
        if (!strcmp(params.param_str.data(), "NV_MEMORY_PROFILER")) {
@@ -48,8 +48,8 @@ u32 nvhost_ctrl::NvOsGetConfigU32(const std::vector<u8>& input, std::vector<u8>&
 u32 nvhost_ctrl::IocCtrlEventWait(const std::vector<u8>& input, std::vector<u8>& output) {
    IocCtrlEventWaitParams params{};
    std::memcpy(&params, input.data(), sizeof(params));
-    LOG_WARNING(Service_NVDRV, "(STUBBED) called, syncpt_id=%u threshold=%u timeout=%d",
+    NGLOG_WARNING(Service_NVDRV, "(STUBBED) called, syncpt_id={} threshold={} timeout={}",
-                params.syncpt_id, params.threshold, params.timeout);
+                  params.syncpt_id, params.threshold, params.timeout);
    // TODO(Subv): Implement actual syncpt waiting.
    params.value = 0;
diff --git a/src/core/hle/service/nvdrv/devices/nvhost_ctrl_gpu.cpp b/src/core/hle/service/nvdrv/devices/nvhost_ctrl_gpu.cpp
index 5008d7cbf..3c78ecaea 100644
--- a/src/core/hle/service/nvdrv/devices/nvhost_ctrl_gpu.cpp
+++ b/src/core/hle/service/nvdrv/devices/nvhost_ctrl_gpu.cpp
@@ -10,8 +10,8 @@
 namespace Service::Nvidia::Devices {
 u32 nvhost_ctrl_gpu::ioctl(Ioctl command, const std::vector<u8>& input, std::vector<u8>& output) {
-    LOG_DEBUG(Service_NVDRV, "called, command=0x%08x, input_size=0x%zx, output_size=0x%zx",
+    NGLOG_DEBUG(Service_NVDRV, "called, command={:#010X}, input_size={:#X}, output_size={:#X}",
-              command.raw, input.size(), output.size());
+                command.raw, input.size(), output.size());
    switch (static_cast<IoctlCommand>(command.raw)) {
    case IoctlCommand::IocGetCharacteristicsCommand:
@@ -25,12 +25,12 @@ u32 nvhost_ctrl_gpu::ioctl(Ioctl command, const std::vector<u8>& input, std::vec
    case IoctlCommand::IocZcullGetInfo:
        return ZCullGetInfo(input, output);
    }
-    UNIMPLEMENTED();
+    UNIMPLEMENTED_MSG("Unimplemented ioctl");
    return 0;
 }
 u32 nvhost_ctrl_gpu::GetCharacteristics(const std::vector<u8>& input, std::vector<u8>& output) {
-    LOG_DEBUG(Service_NVDRV, "called");
+    NGLOG_DEBUG(Service_NVDRV, "called");
    IoctlCharacteristics params{};
    std::memcpy(&params, input.data(), input.size());
    params.gc.arch = 0x120;
@@ -77,15 +77,15 @@ u32 nvhost_ctrl_gpu::GetCharacteristics(const std::vector<u8>& input, std::vecto
 u32 nvhost_ctrl_gpu::GetTPCMasks(const std::vector<u8>& input, std::vector<u8>& output) {
    IoctlGpuGetTpcMasksArgs params{};
    std::memcpy(&params, input.data(), input.size());
-    LOG_WARNING(Service_NVDRV, "(STUBBED) called, mask=0x%x, mask_buf_addr=0x%" PRIx64,
+    NGLOG_WARNING(Service_NVDRV, "(STUBBED) called, mask={:#X}, mask_buf_addr={:#X}",
-                params.mask_buf_size, params.mask_buf_addr);
+                  params.mask_buf_size, params.mask_buf_addr);
    params.unk = 0xcafe; // TODO(ogniK): Needs to be non 0, what does this actually do?
    std::memcpy(output.data(), &params, sizeof(params));
    return 0;
 }
 u32 nvhost_ctrl_gpu::GetActiveSlotMask(const std::vector<u8>& input, std::vector<u8>& output) {
-    LOG_DEBUG(Service_NVDRV, "called");
+    NGLOG_DEBUG(Service_NVDRV, "called");
    IoctlActiveSlotMask params{};
    std::memcpy(&params, input.data(), input.size());
    params.slot = 0x07;
@@ -95,7 +95,7 @@ u32 nvhost_ctrl_gpu::GetActiveSlotMask(const std::vector<u8>& input, std::vector
 }
 u32 nvhost_ctrl_gpu::ZCullGetCtxSize(const std::vector<u8>& input, std::vector<u8>& output) {
-    LOG_DEBUG(Service_NVDRV, "called");
+    NGLOG_DEBUG(Service_NVDRV, "called");
    IoctlZcullGetCtxSize params{};
    std::memcpy(&params, input.data(), input.size());
    params.size = 0x1;
@@ -104,7 +104,7 @@ u32 nvhost_ctrl_gpu::ZCullGetCtxSize(const std::vector<u8>& input, std::vector<u
 }
 u32 nvhost_ctrl_gpu::ZCullGetInfo(const std::vector<u8>& input, std::vector<u8>& output) {
-    LOG_DEBUG(Service_NVDRV, "called");
+    NGLOG_DEBUG(Service_NVDRV, "called");
    IoctlNvgpuGpuZcullGetInfoArgs params{};
    std::memcpy(&params, input.data(), input.size());
    params.width_align_pixels = 0x20;
diff --git a/src/core/hle/service/nvdrv/devices/nvhost_gpu.cpp b/src/core/hle/service/nvdrv/devices/nvhost_gpu.cpp
index a16e90457..70625211e 100644
--- a/src/core/hle/service/nvdrv/devices/nvhost_gpu.cpp
+++ b/src/core/hle/service/nvdrv/devices/nvhost_gpu.cpp
@@ -12,8 +12,8 @@
 namespace Service::Nvidia::Devices {
 u32 nvhost_gpu::ioctl(Ioctl command, const std::vector<u8>& input, std::vector<u8>& output) {
-    LOG_DEBUG(Service_NVDRV, "called, command=0x%08x, input_size=0x%zx, output_size=0x%zx",
+    NGLOG_DEBUG(Service_NVDRV, "called, command={:#010X}, input_size={:#X}, output_size={:#X}",
-              command.raw, input.size(), output.size());
+                command.raw, input.size(), output.size());
    switch (static_cast<IoctlCommand>(command.raw)) {
    case IoctlCommand::IocSetNVMAPfdCommand:
@@ -40,21 +40,21 @@ u32 nvhost_gpu::ioctl(Ioctl command, const std::vector<u8>& input, std::vector<u
        }
    }
-    UNIMPLEMENTED();
+    UNIMPLEMENTED_MSG("Unimplemented ioctl");
    return 0;
 };
 u32 nvhost_gpu::SetNVMAPfd(const std::vector<u8>& input, std::vector<u8>& output) {
    IoctlSetNvmapFD params{};
    std::memcpy(&params, input.data(), input.size());
-    LOG_DEBUG(Service_NVDRV, "called, fd=%x", params.nvmap_fd);
+    NGLOG_DEBUG(Service_NVDRV, "called, fd={}", params.nvmap_fd);
    nvmap_fd = params.nvmap_fd;
    std::memcpy(output.data(), &params, output.size());
    return 0;
 }
 u32 nvhost_gpu::SetClientData(const std::vector<u8>& input, std::vector<u8>& output) {
-    LOG_DEBUG(Service_NVDRV, "called");
+    NGLOG_DEBUG(Service_NVDRV, "called");
    IoctlClientData params{};
    std::memcpy(&params, input.data(), input.size());
    user_data = params.data;
@@ -63,7 +63,7 @@ u32 nvhost_gpu::SetClientData(const std::vector<u8>& input, std::vector<u8>& out
 }
 u32 nvhost_gpu::GetClientData(const std::vector<u8>& input, std::vector<u8>& output) {
-    LOG_DEBUG(Service_NVDRV, "called");
+    NGLOG_DEBUG(Service_NVDRV, "called");
    IoctlClientData params{};
    std::memcpy(&params, input.data(), input.size());
    params.data = user_data;
@@ -73,8 +73,8 @@ u32 nvhost_gpu::GetClientData(const std::vector<u8>& input, std::vector<u8>& out
 u32 nvhost_gpu::ZCullBind(const std::vector<u8>& input, std::vector<u8>& output) {
    std::memcpy(&zcull_params, input.data(), input.size());
-    LOG_DEBUG(Service_NVDRV, "called, gpu_va=%" PRIx64 ", mode=%x", zcull_params.gpu_va,
+    NGLOG_DEBUG(Service_NVDRV, "called, gpu_va={:X}, mode={:X}", zcull_params.gpu_va,
-              zcull_params.mode);
+                zcull_params.mode);
    std::memcpy(output.data(), &zcull_params, output.size());
    return 0;
 }
@@ -82,15 +82,15 @@ u32 nvhost_gpu::ZCullBind(const std::vector<u8>& input, std::vector<u8>& output)
 u32 nvhost_gpu::SetErrorNotifier(const std::vector<u8>& input, std::vector<u8>& output) {
    IoctlSetErrorNotifier params{};
    std::memcpy(&params, input.data(), input.size());
-    LOG_WARNING(Service_NVDRV, "(STUBBED) called, offset=%" PRIx64 ", size=%" PRIx64 ", mem=%x",
+    NGLOG_WARNING(Service_NVDRV, "(STUBBED) called, offset={:X}, size={:X}, mem={:X}",
-                params.offset, params.size, params.mem);
+                  params.offset, params.size, params.mem);
    std::memcpy(output.data(), &params, output.size());
    return 0;
 }
 u32 nvhost_gpu::SetChannelPriority(const std::vector<u8>& input, std::vector<u8>& output) {
    std::memcpy(&channel_priority, input.data(), input.size());
-    LOG_DEBUG(Service_NVDRV, "(STUBBED) called, priority=%x", channel_priority);
+    NGLOG_DEBUG(Service_NVDRV, "(STUBBED) called, priority={:X}", channel_priority);
    std::memcpy(output.data(), &channel_priority, output.size());
    return 0;
 }
@@ -98,10 +98,11 @@ u32 nvhost_gpu::SetChannelPriority(const std::vector<u8>& input, std::vector<u8>
 u32 nvhost_gpu::AllocGPFIFOEx2(const std::vector<u8>& input, std::vector<u8>& output) {
    IoctlAllocGpfifoEx2 params{};
    std::memcpy(&params, input.data(), input.size());
-    LOG_WARNING(Service_NVDRV,
+    NGLOG_WARNING(Service_NVDRV,
-                "(STUBBED) called, num_entries=%x, flags=%x, unk0=%x, unk1=%x, unk2=%x, unk3=%x",
+                  "(STUBBED) called, num_entries={:X}, flags={:X}, unk0={:X}, "
-                params.num_entries, params.flags, params.unk0, params.unk1, params.unk2,
+                  "unk1={:X}, unk2={:X}, unk3={:X}",
-                params.unk3);
+                  params.num_entries, params.flags, params.unk0, params.unk1, params.unk2,
+                  params.unk3);
    params.fence_out.id = 0;
    params.fence_out.value = 0;
    std::memcpy(output.data(), &params, output.size());
@@ -111,8 +112,8 @@ u32 nvhost_gpu::AllocGPFIFOEx2(const std::vector<u8>& input, std::vector<u8>& ou
 u32 nvhost_gpu::AllocateObjectContext(const std::vector<u8>& input, std::vector<u8>& output) {
    IoctlAllocObjCtx params{};
    std::memcpy(&params, input.data(), input.size());
-    LOG_WARNING(Service_NVDRV, "(STUBBED) called, class_num=%x, flags=%x", params.class_num,
+    NGLOG_WARNING(Service_NVDRV, "(STUBBED) called, class_num={:X}, flags={:X}", params.class_num,
-                params.flags);
+                  params.flags);
    params.obj_id = 0x0;
    std::memcpy(output.data(), &params, output.size());
    return 0;
@@ -123,8 +124,8 @@ u32 nvhost_gpu::SubmitGPFIFO(const std::vector<u8>& input, std::vector<u8>& outp
        UNIMPLEMENTED();
    IoctlSubmitGpfifo params{};
    std::memcpy(&params, input.data(), sizeof(IoctlSubmitGpfifo));
-    LOG_WARNING(Service_NVDRV, "(STUBBED) called, gpfifo=%" PRIx64 ", num_entries=%x, flags=%x",
+    NGLOG_WARNING(Service_NVDRV, "(STUBBED) called, gpfifo={:X}, num_entries={:X}, flags={:X}",
-                params.gpfifo, params.num_entries, params.flags);
+                  params.gpfifo, params.num_entries, params.flags);
    auto entries = std::vector<IoctlGpfifoEntry>();
    entries.resize(params.num_entries);
diff --git a/src/core/hle/service/nvdrv/devices/nvmap.cpp b/src/core/hle/service/nvdrv/devices/nvmap.cpp
index 4bb1f57f6..11df8849d 100644
--- a/src/core/hle/service/nvdrv/devices/nvmap.cpp
+++ b/src/core/hle/service/nvdrv/devices/nvmap.cpp
@@ -32,7 +32,7 @@ u32 nvmap::ioctl(Ioctl command, const std::vector<u8>& input, std::vector<u8>& o
        return IocParam(input, output);
    }
-    UNIMPLEMENTED();
+    UNIMPLEMENTED_MSG("Unimplemented ioctl");
    return 0;
 }
@@ -49,7 +49,7 @@ u32 nvmap::IocCreate(const std::vector<u8>& input, std::vector<u8>& output) {
    u32 handle = next_handle++;
    handles[handle] = std::move(object);
-    LOG_DEBUG(Service_NVDRV, "size=0x%08X", params.size);
+    NGLOG_DEBUG(Service_NVDRV, "size={:#010X}", params.size);
    params.handle = handle;
@@ -70,7 +70,7 @@ u32 nvmap::IocAlloc(const std::vector<u8>& input, std::vector<u8>& output) {
    object->addr = params.addr;
    object->status = Object::Status::Allocated;
-    LOG_DEBUG(Service_NVDRV, "called, addr=0x%" PRIx64, params.addr);
+    NGLOG_DEBUG(Service_NVDRV, "called, addr={:X}", params.addr);
    std::memcpy(output.data(), &params, sizeof(params));
    return 0;
@@ -80,7 +80,7 @@ u32 nvmap::IocGetId(const std::vector<u8>& input, std::vector<u8>& output) {
    IocGetIdParams params;
    std::memcpy(&params, input.data(), sizeof(params));
-    LOG_WARNING(Service_NVDRV, "called");
+    NGLOG_WARNING(Service_NVDRV, "called");
    auto object = GetObject(params.handle);
    ASSERT(object);
@@ -95,7 +95,7 @@ u32 nvmap::IocFromId(const std::vector<u8>& input, std::vector<u8>& output) {
    IocFromIdParams params;
    std::memcpy(&params, input.data(), sizeof(params));
-    LOG_WARNING(Service_NVDRV, "(STUBBED) called");
+    NGLOG_WARNING(Service_NVDRV, "(STUBBED) called");
    auto itr = std::find_if(handles.begin(), handles.end(),
                            [&](const auto& entry) { return entry.second->id == params.id; });
@@ -114,7 +114,7 @@ u32 nvmap::IocParam(const std::vector<u8>& input, std::vector<u8>& output) {
    IocParamParams params;
    std::memcpy(&params, input.data(), sizeof(params));
-    LOG_WARNING(Service_NVDRV, "(STUBBED) called type=%u", params.type);
+    NGLOG_WARNING(Service_NVDRV, "(STUBBED) called type={}", params.type);
    auto object = GetObject(params.handle);
    ASSERT(object);
diff --git a/src/core/hle/service/nvdrv/interface.cpp b/src/core/hle/service/nvdrv/interface.cpp
index d0d64a840..38b3a9a84 100644
--- a/src/core/hle/service/nvdrv/interface.cpp
+++ b/src/core/hle/service/nvdrv/interface.cpp
@@ -12,7 +12,7 @@
 namespace Service::Nvidia {
 void NVDRV::Open(Kernel::HLERequestContext& ctx) {
-    LOG_DEBUG(Service_NVDRV, "called");
+    NGLOG_DEBUG(Service_NVDRV, "called");
    const auto& buffer = ctx.ReadBuffer();
    std::string device_name(buffer.begin(), buffer.end());
@@ -25,7 +25,7 @@ void NVDRV::Open(Kernel::HLERequestContext& ctx) {
 }
 void NVDRV::Ioctl(Kernel::HLERequestContext& ctx) {
-    LOG_DEBUG(Service_NVDRV, "called");
+    NGLOG_DEBUG(Service_NVDRV, "called");
    IPC::RequestParser rp{ctx};
    u32 fd = rp.Pop<u32>();
@@ -41,7 +41,7 @@ void NVDRV::Ioctl(Kernel::HLERequestContext& ctx) {
 }
 void NVDRV::Close(Kernel::HLERequestContext& ctx) {
-    LOG_DEBUG(Service_NVDRV, "called");
+    NGLOG_DEBUG(Service_NVDRV, "called");
    IPC::RequestParser rp{ctx};
    u32 fd = rp.Pop<u32>();
@@ -53,7 +53,7 @@ void NVDRV::Close(Kernel::HLERequestContext& ctx) {
 }
 void NVDRV::Initialize(Kernel::HLERequestContext& ctx) {
-    LOG_WARNING(Service_NVDRV, "(STUBBED) called");
+    NGLOG_WARNING(Service_NVDRV, "(STUBBED) called");
    IPC::ResponseBuilder rb{ctx, 3};
    rb.Push(RESULT_SUCCESS);
    rb.Push<u32>(0);
@@ -63,7 +63,7 @@ void NVDRV::QueryEvent(Kernel::HLERequestContext& ctx) {
    IPC::RequestParser rp{ctx};
    u32 fd = rp.Pop<u32>();
    u32 event_id = rp.Pop<u32>();
-    LOG_WARNING(Service_NVDRV, "(STUBBED) called, fd=%x, event_id=%x", fd, event_id);
+    NGLOG_WARNING(Service_NVDRV, "(STUBBED) called, fd={:X}, event_id={:X}", fd, event_id);
    IPC::ResponseBuilder rb{ctx, 3, 1};
    rb.Push(RESULT_SUCCESS);
@@ -75,14 +75,14 @@ void NVDRV::SetClientPID(Kernel::HLERequestContext& ctx) {
    IPC::RequestParser rp{ctx};
    pid = rp.Pop<u64>();
-    LOG_WARNING(Service_NVDRV, "(STUBBED) called, pid=0x%" PRIx64, pid);
+    NGLOG_WARNING(Service_NVDRV, "(STUBBED) called, pid={:#X}", pid);
    IPC::ResponseBuilder rb{ctx, 3};
    rb.Push(RESULT_SUCCESS);
    rb.Push<u32>(0);
 }
 void NVDRV::FinishInitialize(Kernel::HLERequestContext& ctx) {
-    LOG_WARNING(Service_NVDRV, "(STUBBED) called");
+    NGLOG_WARNING(Service_NVDRV, "(STUBBED) called");
    IPC::ResponseBuilder rb{ctx, 2};
    rb.Push(RESULT_SUCCESS);
 }
diff --git a/src/core/hle/service/nvflinger/buffer_queue.cpp b/src/core/hle/service/nvflinger/buffer_queue.cpp
index 03a4fed59..49e88b394 100644
--- a/src/core/hle/service/nvflinger/buffer_queue.cpp
+++ b/src/core/hle/service/nvflinger/buffer_queue.cpp
@@ -9,7 +9,8 @@
 #include "core/core_timing.h"
 #include "core/hle/service/nvflinger/buffer_queue.h"
-namespace Service::NVFlinger {
+namespace Service {
+namespace NVFlinger {
 BufferQueue::BufferQueue(u32 id, u64 layer_id) : id(id), layer_id(layer_id) {
    native_handle = Kernel::Event::Create(Kernel::ResetType::OneShot, "BufferQueue NativeHandle");
@@ -22,7 +23,7 @@ void BufferQueue::SetPreallocatedBuffer(u32 slot, IGBPBuffer& igbp_buffer) {
    buffer.igbp_buffer = igbp_buffer;
    buffer.status = Buffer::Status::Free;
-    LOG_WARNING(Service, "Adding graphics buffer %u", slot);
+    NGLOG_WARNING(Service, "Adding graphics buffer {}", slot);
    queue.emplace_back(buffer);
@@ -93,7 +94,7 @@ void BufferQueue::ReleaseBuffer(u32 slot) {
 }
 u32 BufferQueue::Query(QueryType type) {
-    LOG_WARNING(Service, "(STUBBED) called type=%u", static_cast<u32>(type));
+    NGLOG_WARNING(Service, "(STUBBED) called type={}", static_cast<u32>(type));
    switch (type) {
    case QueryType::NativeWindowFormat:
        // TODO(Subv): Use an enum for this
@@ -110,4 +111,5 @@ void BufferQueue::SetBufferWaitEvent(Kernel::SharedPtr<Kernel::Event>&& wait_eve
    buffer_wait_event = std::move(wait_event);
 }
-} // namespace Service::NVFlinger
+} // namespace NVFlinger
+} // namespace Service
diff --git a/src/core/hle/service/nvflinger/buffer_queue.h b/src/core/hle/service/nvflinger/buffer_queue.h
index 95adc4706..1de5767cb 100644
--- a/src/core/hle/service/nvflinger/buffer_queue.h
+++ b/src/core/hle/service/nvflinger/buffer_queue.h
@@ -13,7 +13,8 @@ namespace CoreTiming {
 struct EventType;
 }
-namespace Service::NVFlinger {
+namespace Service {
+namespace NVFlinger {
 struct IGBPBuffer {
    u32_le magic;
@@ -97,4 +98,5 @@ private:
    Kernel::SharedPtr<Kernel::Event> buffer_wait_event;
 };
-} // namespace Service::NVFlinger
+} // namespace NVFlinger
+} // namespace Service
diff --git a/src/core/hle/service/nvflinger/nvflinger.cpp b/src/core/hle/service/nvflinger/nvflinger.cpp
index a99ebc8e6..3481e48d0 100644
--- a/src/core/hle/service/nvflinger/nvflinger.cpp
+++ b/src/core/hle/service/nvflinger/nvflinger.cpp
@@ -48,7 +48,7 @@ NVFlinger::~NVFlinger() {
 }
 u64 NVFlinger::OpenDisplay(const std::string& name) {
-    LOG_WARNING(Service, "Opening display %s", name.c_str());
+    NGLOG_WARNING(Service, "Opening display {}", name);
    // TODO(Subv): Currently we only support the Default display.
    ASSERT(name == "Default");
diff --git a/src/core/hle/service/pctl/pctl_a.cpp b/src/core/hle/service/pctl/module.cpp
index 9fb4628ad..dd20d5ae7 100644
--- a/src/core/hle/service/pctl/pctl_a.cpp
+++ b/src/core/hle/service/pctl/module.cpp
@@ -4,7 +4,8 @@
 #include "common/logging/log.h"
 #include "core/hle/ipc_helpers.h"
-#include "core/hle/service/pctl/pctl_a.h"
+#include "core/hle/service/pctl/module.h"
+#include "core/hle/service/pctl/pctl.h"
 namespace Service::PCTL {
@@ -12,7 +13,7 @@ class IParentalControlService final : public ServiceFramework<IParentalControlSe
 public:
    IParentalControlService() : ServiceFramework("IParentalControlService") {
        static const FunctionInfo functions[] = {
-            {1, nullptr, "Initialize"},
+            {1, &IParentalControlService::Initialize, "Initialize"},
            {1001, nullptr, "CheckFreeCommunicationPermission"},
            {1002, nullptr, "ConfirmLaunchApplicationPermission"},
            {1003, nullptr, "ConfirmResumeApplicationPermission"},
@@ -108,20 +109,38 @@ public:
        };
        RegisterHandlers(functions);
    }
+private:
+    void Initialize(Kernel::HLERequestContext& ctx) {
+        NGLOG_WARNING(Service_PCTL, "(STUBBED) called");
+        IPC::ResponseBuilder rb{ctx, 2, 0, 0};
+        rb.Push(RESULT_SUCCESS);
+    }
 };
-void PCTL_A::CreateService(Kernel::HLERequestContext& ctx) {
+void Module::Interface::CreateService(Kernel::HLERequestContext& ctx) {
    IPC::ResponseBuilder rb{ctx, 2, 0, 1};
    rb.Push(RESULT_SUCCESS);
    rb.PushIpcInterface<IParentalControlService>();
-    LOG_DEBUG(Service_PCTL, "called");
+    NGLOG_DEBUG(Service_PCTL, "called");
 }
-PCTL_A::PCTL_A() : ServiceFramework("pctl:a") {
+void Module::Interface::CreateServiceWithoutInitialize(Kernel::HLERequestContext& ctx) {
-    static const FunctionInfo functions[] = {
+    IPC::ResponseBuilder rb{ctx, 2, 0, 1};
-        {0, &PCTL_A::CreateService, "CreateService"},
+    rb.Push(RESULT_SUCCESS);
-        {1, nullptr, "CreateServiceWithoutInitialize"},
+    rb.PushIpcInterface<IParentalControlService>();
-    };
+    NGLOG_DEBUG(Service_PCTL, "called");
-    RegisterHandlers(functions);
+}
+Module::Interface::Interface(std::shared_ptr<Module> module, const char* name)
+    : ServiceFramework(name), module(std::move(module)) {}
+void InstallInterfaces(SM::ServiceManager& service_manager) {
+    auto module = std::make_shared<Module>();
+    std::make_shared<PCTL>(module, "pctl")->InstallAsService(service_manager);
+    std::make_shared<PCTL>(module, "pctl:a")->InstallAsService(service_manager);
+    std::make_shared<PCTL>(module, "pctl:r")->InstallAsService(service_manager);
+    std::make_shared<PCTL>(module, "pctl:s")->InstallAsService(service_manager);
 }
 } // namespace Service::PCTL
diff --git a/src/core/hle/service/pctl/module.h b/src/core/hle/service/pctl/module.h
new file mode 100644
index 000000000..68da628a8
--- /dev/null
+++ b/src/core/hle/service/pctl/module.h
@@ -0,0 +1,28 @@
+// Copyright 2018 yuzu emulator team
+// Licensed under GPLv2 or any later version
+// Refer to the license.txt file included.
+#pragma once
+#include "core/hle/service/service.h"
+namespace Service::PCTL {
+class Module final {
+public:
+    class Interface : public ServiceFramework<Interface> {
+    public:
+        Interface(std::shared_ptr<Module> module, const char* name);
+        void CreateService(Kernel::HLERequestContext& ctx);
+        void CreateServiceWithoutInitialize(Kernel::HLERequestContext& ctx);
+    protected:
+        std::shared_ptr<Module> module;
+    };
+};
+/// Registers all PCTL services with the specified service manager.
+void InstallInterfaces(SM::ServiceManager& service_manager);
+} // namespace Service::PCTL
diff --git a/src/core/hle/service/pctl/pctl.cpp b/src/core/hle/service/pctl/pctl.cpp
index 6ee81866d..de2741d66 100644
--- a/src/core/hle/service/pctl/pctl.cpp
+++ b/src/core/hle/service/pctl/pctl.cpp
@@ -3,12 +3,15 @@
 // Refer to the license.txt file included.
 #include "core/hle/service/pctl/pctl.h"
-#include "core/hle/service/pctl/pctl_a.h"
 namespace Service::PCTL {
-void InstallInterfaces(SM::ServiceManager& service_manager) {
+PCTL::PCTL(std::shared_ptr<Module> module, const char* name)
-    std::make_shared<PCTL_A>()->InstallAsService(service_manager);
+    : Module::Interface(std::move(module), name) {
+    static const FunctionInfo functions[] = {
+        {0, &PCTL::CreateService, "CreateService"},
+        {1, &PCTL::CreateServiceWithoutInitialize, "CreateServiceWithoutInitialize"},
+    };
+    RegisterHandlers(functions);
 }
 } // namespace Service::PCTL
diff --git a/src/core/hle/service/pctl/pctl.h b/src/core/hle/service/pctl/pctl.h
index f0a84b115..8ddf69128 100644
--- a/src/core/hle/service/pctl/pctl.h
+++ b/src/core/hle/service/pctl/pctl.h
@@ -4,11 +4,13 @@
 #pragma once
-#include "core/hle/service/service.h"
+#include "core/hle/service/pctl/module.h"
 namespace Service::PCTL {
-/// Registers all PCTL services with the specified service manager.
+class PCTL final : public Module::Interface {
-void InstallInterfaces(SM::ServiceManager& service_manager);
+public:
+    explicit PCTL(std::shared_ptr<Module> module, const char* name);
+};
 } // namespace Service::PCTL
diff --git a/src/core/hle/service/pctl/pctl_a.h b/src/core/hle/service/pctl/pctl_a.h
deleted file mode 100644
index 09ed82e1b..000000000
--- a/src/core/hle/service/pctl/pctl_a.h
+++ /dev/null
@@ -1,20 +0,0 @@
-// Copyright 2018 yuzu emulator team
-// Licensed under GPLv2 or any later version
-// Refer to the license.txt file included.
-#pragma once
-#include "core/hle/service/service.h"
-namespace Service::PCTL {
-class PCTL_A final : public ServiceFramework<PCTL_A> {
-public:
-    PCTL_A();
-    ~PCTL_A() = default;
-private:
-    void CreateService(Kernel::HLERequestContext& ctx);
-};
-} // namespace Service::PCTL
diff --git a/src/core/hle/service/service.cpp b/src/core/hle/service/service.cpp
index 68d2b9f17..34d691b90 100644
--- a/src/core/hle/service/service.cpp
+++ b/src/core/hle/service/service.cpp
@@ -121,7 +121,7 @@ void ServiceFrameworkBase::ReportUnimplementedFunction(Kernel::HLERequestContext
    }
    buf.push_back('}');
-    LOG_ERROR(Service, "unknown / unimplemented %s", fmt::to_string(buf).c_str());
+    NGLOG_ERROR(Service, "unknown / unimplemented {}", fmt::to_string(buf));
    UNIMPLEMENTED();
 }
@@ -132,8 +132,8 @@ void ServiceFrameworkBase::InvokeRequest(Kernel::HLERequestContext& ctx) {
        return ReportUnimplementedFunction(ctx, info);
    }
-    LOG_TRACE(
+    NGLOG_TRACE(
-        Service, "%s",
+        Service, "{}",
        MakeFunctionString(info->name, GetServiceName().c_str(), ctx.CommandBuffer()).c_str());
    handler_invoker(this, info->handler_callback, ctx);
 }
@@ -201,12 +201,12 @@ void Init(std::shared_ptr<SM::ServiceManager>& sm) {
    VI::InstallInterfaces(*sm, nv_flinger);
    Set::InstallInterfaces(*sm);
-    LOG_DEBUG(Service, "initialized OK");
+    NGLOG_DEBUG(Service, "initialized OK");
 }
 /// Shutdown ServiceManager
 void Shutdown() {
    g_kernel_named_ports.clear();
-    LOG_DEBUG(Service, "shutdown OK");
+    NGLOG_DEBUG(Service, "shutdown OK");
 }
 } // namespace Service
diff --git a/src/core/hle/service/set/set.cpp b/src/core/hle/service/set/set.cpp
index fc3e424d0..ece29aa70 100644
--- a/src/core/hle/service/set/set.cpp
+++ b/src/core/hle/service/set/set.cpp
@@ -22,7 +22,7 @@ void SET::GetAvailableLanguageCodes(Kernel::HLERequestContext& ctx) {
    rb.Push(RESULT_SUCCESS);
-    LOG_WARNING(Service_SET, "(STUBBED) called");
+    NGLOG_WARNING(Service_SET, "(STUBBED) called");
 }
 SET::SET() : ServiceFramework("set") {
diff --git a/src/core/hle/service/set/set_sys.cpp b/src/core/hle/service/set/set_sys.cpp
index fa85277fe..762a664c5 100644
--- a/src/core/hle/service/set/set_sys.cpp
+++ b/src/core/hle/service/set/set_sys.cpp
@@ -16,7 +16,7 @@ void SET_SYS::GetColorSetId(Kernel::HLERequestContext& ctx) {
    rb.Push(RESULT_SUCCESS);
    rb.Push<u32>(0);
-    LOG_WARNING(Service_SET, "(STUBBED) called");
+    NGLOG_WARNING(Service_SET, "(STUBBED) called");
 }
 SET_SYS::SET_SYS() : ServiceFramework("set:sys") {
diff --git a/src/core/hle/service/sm/controller.cpp b/src/core/hle/service/sm/controller.cpp
index 13e31620d..fe5097cdc 100644
--- a/src/core/hle/service/sm/controller.cpp
+++ b/src/core/hle/service/sm/controller.cpp
@@ -17,7 +17,7 @@ void Controller::ConvertSessionToDomain(Kernel::HLERequestContext& ctx) {
    rb.Push(RESULT_SUCCESS);
    rb.Push<u32>(1); // Converted sessions start with 1 request handler
-    LOG_DEBUG(Service, "called, server_session=%d", ctx.Session()->GetObjectId());
+    NGLOG_DEBUG(Service, "called, server_session={}", ctx.Session()->GetObjectId());
 }
 void Controller::DuplicateSession(Kernel::HLERequestContext& ctx) {
@@ -29,11 +29,11 @@ void Controller::DuplicateSession(Kernel::HLERequestContext& ctx) {
    Kernel::SharedPtr<Kernel::ClientSession> session{ctx.Session()->parent->client};
    rb.PushMoveObjects(session);
-    LOG_DEBUG(Service, "called, session=%u", session->GetObjectId());
+    NGLOG_DEBUG(Service, "called, session={}", session->GetObjectId());
 }
 void Controller::DuplicateSessionEx(Kernel::HLERequestContext& ctx) {
-    LOG_WARNING(Service, "(STUBBED) called, using DuplicateSession");
+    NGLOG_WARNING(Service, "(STUBBED) called, using DuplicateSession");
    DuplicateSession(ctx);
 }
@@ -43,7 +43,7 @@ void Controller::QueryPointerBufferSize(Kernel::HLERequestContext& ctx) {
    rb.Push(RESULT_SUCCESS);
    rb.Push<u32>(0x500);
-    LOG_WARNING(Service, "(STUBBED) called");
+    NGLOG_WARNING(Service, "(STUBBED) called");
 }
 Controller::Controller() : ServiceFramework("IpcController") {
diff --git a/src/core/hle/service/sm/sm.cpp b/src/core/hle/service/sm/sm.cpp
index 4578fc05f..073277ade 100644
--- a/src/core/hle/service/sm/sm.cpp
+++ b/src/core/hle/service/sm/sm.cpp
@@ -86,7 +86,7 @@ SM::~SM() = default;
 void SM::Initialize(Kernel::HLERequestContext& ctx) {
    IPC::ResponseBuilder rb{ctx, 2};
    rb.Push(RESULT_SUCCESS);
-    LOG_DEBUG(Service_SM, "called");
+    NGLOG_DEBUG(Service_SM, "called");
 }
 void SM::GetService(Kernel::HLERequestContext& ctx) {
@@ -102,8 +102,8 @@ void SM::GetService(Kernel::HLERequestContext& ctx) {
    if (client_port.Failed()) {
        IPC::ResponseBuilder rb = rp.MakeBuilder(2, 0, 0);
        rb.Push(client_port.Code());
-        LOG_ERROR(Service_SM, "called service=%s -> error 0x%08X", name.c_str(),
+        NGLOG_ERROR(Service_SM, "called service={} -> error {:#010X}", name,
-                  client_port.Code().raw);
+                    client_port.Code().raw);
        if (name.length() == 0)
            return; // LibNX Fix
        UNIMPLEMENTED();
@@ -113,8 +113,7 @@ void SM::GetService(Kernel::HLERequestContext& ctx) {
    auto session = client_port.Unwrap()->Connect();
    ASSERT(session.Succeeded());
    if (session.Succeeded()) {
-        LOG_DEBUG(Service_SM, "called service=%s -> session=%u", name.c_str(),
+        NGLOG_DEBUG(Service_SM, "called service={} -> session={}", name, (*session)->GetObjectId());
-                  (*session)->GetObjectId());
        IPC::ResponseBuilder rb =
            rp.MakeBuilder(2, 0, 1, IPC::ResponseBuilder::Flags::AlwaysMoveHandles);
        rb.Push(session.Code());
diff --git a/src/core/hle/service/sockets/bsd.cpp b/src/core/hle/service/sockets/bsd.cpp
index f99809bed..ab909fdaa 100644
--- a/src/core/hle/service/sockets/bsd.cpp
+++ b/src/core/hle/service/sockets/bsd.cpp
@@ -8,7 +8,7 @@
 namespace Service::Sockets {
 void BSD::RegisterClient(Kernel::HLERequestContext& ctx) {
-    LOG_WARNING(Service, "(STUBBED) called");
+    NGLOG_WARNING(Service, "(STUBBED) called");
    IPC::ResponseBuilder rb{ctx, 3};
@@ -17,7 +17,7 @@ void BSD::RegisterClient(Kernel::HLERequestContext& ctx) {
 }
 void BSD::StartMonitoring(Kernel::HLERequestContext& ctx) {
-    LOG_WARNING(Service, "(STUBBED) called");
+    NGLOG_WARNING(Service, "(STUBBED) called");
    IPC::ResponseBuilder rb{ctx, 3};
@@ -32,7 +32,8 @@ void BSD::Socket(Kernel::HLERequestContext& ctx) {
    u32 type = rp.Pop<u32>();
    u32 protocol = rp.Pop<u32>();
-    LOG_WARNING(Service, "(STUBBED) called domain=%u type=%u protocol=%u", domain, type, protocol);
+    NGLOG_WARNING(Service, "(STUBBED) called domain={} type={} protocol={}", domain, type,
+                  protocol);
    u32 fd = next_fd++;
@@ -44,7 +45,7 @@ void BSD::Socket(Kernel::HLERequestContext& ctx) {
 }
 void BSD::Connect(Kernel::HLERequestContext& ctx) {
-    LOG_WARNING(Service, "(STUBBED) called");
+    NGLOG_WARNING(Service, "(STUBBED) called");
    IPC::ResponseBuilder rb{ctx, 4};
@@ -54,7 +55,7 @@ void BSD::Connect(Kernel::HLERequestContext& ctx) {
 }
 void BSD::SendTo(Kernel::HLERequestContext& ctx) {
-    LOG_WARNING(Service, "(STUBBED) called");
+    NGLOG_WARNING(Service, "(STUBBED) called");
    IPC::ResponseBuilder rb{ctx, 4};
@@ -64,7 +65,7 @@ void BSD::SendTo(Kernel::HLERequestContext& ctx) {
 }
 void BSD::Close(Kernel::HLERequestContext& ctx) {
-    LOG_WARNING(Service, "(STUBBED) called");
+    NGLOG_WARNING(Service, "(STUBBED) called");
    IPC::ResponseBuilder rb{ctx, 4};
diff --git a/src/core/hle/service/sockets/sfdnsres.cpp b/src/core/hle/service/sockets/sfdnsres.cpp
index d235c4cfd..f377e59f2 100644
--- a/src/core/hle/service/sockets/sfdnsres.cpp
+++ b/src/core/hle/service/sockets/sfdnsres.cpp
@@ -10,7 +10,7 @@ namespace Service::Sockets {
 void SFDNSRES::GetAddrInfo(Kernel::HLERequestContext& ctx) {
    IPC::RequestParser rp{ctx};
-    LOG_WARNING(Service, "(STUBBED) called");
+    NGLOG_WARNING(Service, "(STUBBED) called");
    IPC::ResponseBuilder rb{ctx, 2};
diff --git a/src/core/hle/service/spl/module.cpp b/src/core/hle/service/spl/module.cpp
index 3f5a342a7..76ba97156 100644
--- a/src/core/hle/service/spl/module.cpp
+++ b/src/core/hle/service/spl/module.cpp
@@ -28,7 +28,7 @@ void Module::Interface::GetRandomBytes(Kernel::HLERequestContext& ctx) {
    IPC::ResponseBuilder rb{ctx, 2};
    rb.Push(RESULT_SUCCESS);
-    LOG_DEBUG(Service_SPL, "called");
+    NGLOG_DEBUG(Service_SPL, "called");
 }
 void InstallInterfaces(SM::ServiceManager& service_manager) {
diff --git a/src/core/hle/service/ssl/ssl.cpp b/src/core/hle/service/ssl/ssl.cpp
index 40aea6090..8a85df91a 100644
--- a/src/core/hle/service/ssl/ssl.cpp
+++ b/src/core/hle/service/ssl/ssl.cpp
@@ -65,7 +65,7 @@ public:
 private:
    void SetOption(Kernel::HLERequestContext& ctx) {
-        LOG_WARNING(Service_SSL, "(STUBBED) called");
+        NGLOG_WARNING(Service_SSL, "(STUBBED) called");
        IPC::RequestParser rp{ctx};
        IPC::ResponseBuilder rb = rp.MakeBuilder(2, 0, 0);
@@ -73,7 +73,7 @@ private:
    }
    void CreateConnection(Kernel::HLERequestContext& ctx) {
-        LOG_WARNING(Service_SSL, "(STUBBED) called");
+        NGLOG_WARNING(Service_SSL, "(STUBBED) called");
        IPC::ResponseBuilder rb{ctx, 2, 0, 1};
        rb.Push(RESULT_SUCCESS);
@@ -82,7 +82,7 @@ private:
 };
 void SSL::CreateContext(Kernel::HLERequestContext& ctx) {
-    LOG_WARNING(Service_SSL, "(STUBBED) called");
+    NGLOG_WARNING(Service_SSL, "(STUBBED) called");
    IPC::ResponseBuilder rb{ctx, 2, 0, 1};
    rb.Push(RESULT_SUCCESS);
diff --git a/src/core/hle/service/time/time.cpp b/src/core/hle/service/time/time.cpp
index 2604ecc1c..278465358 100644
--- a/src/core/hle/service/time/time.cpp
+++ b/src/core/hle/service/time/time.cpp
@@ -32,14 +32,14 @@ private:
        const s64 time_since_epoch{std::chrono::duration_cast<std::chrono::seconds>(
                                       std::chrono::system_clock::now().time_since_epoch())
                                       .count()};
-        LOG_DEBUG(Service_Time, "called");
+        NGLOG_DEBUG(Service_Time, "called");
        IPC::ResponseBuilder rb{ctx, 4};
        rb.Push(RESULT_SUCCESS);
        rb.Push<u64>(time_since_epoch);
    }
    void GetSystemClockContext(Kernel::HLERequestContext& ctx) {
-        LOG_WARNING(Service_Time, "(STUBBED) called");
+        NGLOG_WARNING(Service_Time, "(STUBBED) called");
        SystemClockContext system_clock_ontext{};
        IPC::ResponseBuilder rb{ctx, (sizeof(SystemClockContext) / 4) + 2};
        rb.Push(RESULT_SUCCESS);
@@ -58,7 +58,7 @@ public:
 private:
    void GetCurrentTimePoint(Kernel::HLERequestContext& ctx) {
-        LOG_DEBUG(Service_Time, "called");
+        NGLOG_DEBUG(Service_Time, "called");
        SteadyClockTimePoint steady_clock_time_point{cyclesToMs(CoreTiming::GetTicks()) / 1000};
        IPC::ResponseBuilder rb{ctx, (sizeof(SteadyClockTimePoint) / 4) + 2};
        rb.Push(RESULT_SUCCESS);
@@ -86,7 +86,7 @@ public:
 private:
    void GetDeviceLocationName(Kernel::HLERequestContext& ctx) {
-        LOG_WARNING(Service_Time, "(STUBBED) called");
+        NGLOG_WARNING(Service_Time, "(STUBBED) called");
        LocationName location_name{};
        IPC::ResponseBuilder rb{ctx, (sizeof(LocationName) / 4) + 2};
        rb.Push(RESULT_SUCCESS);
@@ -94,14 +94,14 @@ private:
    }
    void GetTotalLocationNameCount(Kernel::HLERequestContext& ctx) {
-        LOG_WARNING(Service_Time, "(STUBBED) called");
+        NGLOG_WARNING(Service_Time, "(STUBBED) called");
        IPC::ResponseBuilder rb{ctx, 3};
        rb.Push(RESULT_SUCCESS);
        rb.Push<u32>(0);
    }
    void LoadTimeZoneRule(Kernel::HLERequestContext& ctx) {
-        LOG_WARNING(Service_Time, "(STUBBED) called");
+        NGLOG_WARNING(Service_Time, "(STUBBED) called");
        IPC::ResponseBuilder rb{ctx, 2};
        rb.Push(RESULT_SUCCESS);
    }
@@ -110,7 +110,7 @@ private:
        IPC::RequestParser rp{ctx};
        u64 posix_time = rp.Pop<u64>();
-        LOG_WARNING(Service_Time, "(STUBBED) called, posix_time=0x%016lX", posix_time);
+        NGLOG_WARNING(Service_Time, "(STUBBED) called, posix_time={:#018X}", posix_time);
        CalendarTime calendar_time{2018, 1, 1, 0, 0, 0};
        CalendarAdditionalInfo additional_info{};
@@ -125,35 +125,35 @@ void Module::Interface::GetStandardUserSystemClock(Kernel::HLERequestContext& ct
    IPC::ResponseBuilder rb{ctx, 2, 0, 1};
    rb.Push(RESULT_SUCCESS);
    rb.PushIpcInterface<ISystemClock>();
-    LOG_DEBUG(Service_Time, "called");
+    NGLOG_DEBUG(Service_Time, "called");
 }
 void Module::Interface::GetStandardNetworkSystemClock(Kernel::HLERequestContext& ctx) {
    IPC::ResponseBuilder rb{ctx, 2, 0, 1};
    rb.Push(RESULT_SUCCESS);
    rb.PushIpcInterface<ISystemClock>();
-    LOG_DEBUG(Service_Time, "called");
+    NGLOG_DEBUG(Service_Time, "called");
 }
 void Module::Interface::GetStandardSteadyClock(Kernel::HLERequestContext& ctx) {
    IPC::ResponseBuilder rb{ctx, 2, 0, 1};
    rb.Push(RESULT_SUCCESS);
    rb.PushIpcInterface<ISteadyClock>();
-    LOG_DEBUG(Service_Time, "called");
+    NGLOG_DEBUG(Service_Time, "called");
 }
 void Module::Interface::GetTimeZoneService(Kernel::HLERequestContext& ctx) {
    IPC::ResponseBuilder rb{ctx, 2, 0, 1};
    rb.Push(RESULT_SUCCESS);
    rb.PushIpcInterface<ITimeZoneService>();
-    LOG_DEBUG(Service_Time, "called");
+    NGLOG_DEBUG(Service_Time, "called");
 }
 void Module::Interface::GetStandardLocalSystemClock(Kernel::HLERequestContext& ctx) {
    IPC::ResponseBuilder rb{ctx, 2, 0, 1};
    rb.Push(RESULT_SUCCESS);
    rb.PushIpcInterface<ISystemClock>();
-    LOG_DEBUG(Service_Time, "called");
+    NGLOG_DEBUG(Service_Time, "called");
 }
 Module::Interface::Interface(std::shared_ptr<Module> time, const char* name)
diff --git a/src/core/hle/service/vi/vi.cpp b/src/core/hle/service/vi/vi.cpp
index 36ae2215f..45f3568d2 100644
--- a/src/core/hle/service/vi/vi.cpp
+++ b/src/core/hle/service/vi/vi.cpp
@@ -470,7 +470,7 @@ private:
        u32 flags = rp.Pop<u32>();
        auto buffer_queue = nv_flinger->GetBufferQueue(id);
-        LOG_DEBUG(Service_VI, "called, transaction=%x", static_cast<u32>(transaction));
+        NGLOG_DEBUG(Service_VI, "called, transaction={:X}", static_cast<u32>(transaction));
        if (transaction == TransactionId::Connect) {
            IGBPConnectRequestParcel request{ctx.ReadBuffer()};
@@ -532,7 +532,7 @@ private:
            IGBPQueryResponseParcel response{value};
            ctx.WriteBuffer(response.Serialize());
        } else if (transaction == TransactionId::CancelBuffer) {
-            LOG_WARNING(Service_VI, "(STUBBED) called, transaction=CancelBuffer");
+            NGLOG_WARNING(Service_VI, "(STUBBED) called, transaction=CancelBuffer");
        } else {
            ASSERT_MSG(false, "Unimplemented");
        }
@@ -547,7 +547,8 @@ private:
        s32 addval = rp.PopRaw<s32>();
        u32 type = rp.Pop<u32>();
-        LOG_WARNING(Service_VI, "(STUBBED) called id=%u, addval=%08X, type=%08X", id, addval, type);
+        NGLOG_WARNING(Service_VI, "(STUBBED) called id={}, addval={:08X}, type={:08X}", id, addval,
+                      type);
        IPC::ResponseBuilder rb{ctx, 2};
        rb.Push(RESULT_SUCCESS);
    }
@@ -561,7 +562,7 @@ private:
        // TODO(Subv): Find out what this actually is.
-        LOG_WARNING(Service_VI, "(STUBBED) called id=%u, unknown=%08X", id, unknown);
+        NGLOG_WARNING(Service_VI, "(STUBBED) called id={}, unknown={:08X}", id, unknown);
        IPC::ResponseBuilder rb{ctx, 2, 1};
        rb.Push(RESULT_SUCCESS);
        rb.PushCopyObjects(buffer_queue->GetNativeHandle());
@@ -624,7 +625,7 @@ public:
 private:
    void SetLayerZ(Kernel::HLERequestContext& ctx) {
-        LOG_WARNING(Service_VI, "(STUBBED) called");
+        NGLOG_WARNING(Service_VI, "(STUBBED) called");
        IPC::RequestParser rp{ctx};
        u64 layer_id = rp.Pop<u64>();
        u64 z_value = rp.Pop<u64>();
@@ -639,8 +640,8 @@ private:
        bool visibility = rp.Pop<bool>();
        IPC::ResponseBuilder rb = rp.MakeBuilder(2, 0, 0);
        rb.Push(RESULT_SUCCESS);
-        LOG_WARNING(Service_VI, "(STUBBED) called, layer_id=0x%x, visibility=%u", layer_id,
+        NGLOG_WARNING(Service_VI, "(STUBBED) called, layer_id={:#010X}, visibility={}", layer_id,
-                    visibility);
+                      visibility);
    }
 };
@@ -722,7 +723,7 @@ public:
 private:
    void CloseDisplay(Kernel::HLERequestContext& ctx) {
-        LOG_WARNING(Service_VI, "(STUBBED) called");
+        NGLOG_WARNING(Service_VI, "(STUBBED) called");
        IPC::RequestParser rp{ctx};
        u64 display = rp.Pop<u64>();
@@ -731,7 +732,7 @@ private:
    }
    void CreateManagedLayer(Kernel::HLERequestContext& ctx) {
-        LOG_WARNING(Service_VI, "(STUBBED) called");
+        NGLOG_WARNING(Service_VI, "(STUBBED) called");
        IPC::RequestParser rp{ctx};
        u32 unknown = rp.Pop<u32>();
        rp.Skip(1, false);
@@ -746,7 +747,7 @@ private:
    }
    void AddToLayerStack(Kernel::HLERequestContext& ctx) {
-        LOG_WARNING(Service_VI, "(STUBBED) called");
+        NGLOG_WARNING(Service_VI, "(STUBBED) called");
        IPC::RequestParser rp{ctx};
        u32 stack = rp.Pop<u32>();
        u64 layer_id = rp.Pop<u64>();
@@ -761,8 +762,8 @@ private:
        bool visibility = rp.Pop<bool>();
        IPC::ResponseBuilder rb = rp.MakeBuilder(2, 0, 0);
        rb.Push(RESULT_SUCCESS);
-        LOG_WARNING(Service_VI, "(STUBBED) called, layer_id=0x%x, visibility=%u", layer_id,
+        NGLOG_WARNING(Service_VI, "(STUBBED) called, layer_id={:#X}, visibility={}", layer_id,
-                    visibility);
+                      visibility);
    }
    std::shared_ptr<NVFlinger::NVFlinger> nv_flinger;
@@ -775,7 +776,7 @@ public:
 private:
    void GetRelayService(Kernel::HLERequestContext& ctx) {
-        LOG_WARNING(Service_VI, "(STUBBED) called");
+        NGLOG_WARNING(Service_VI, "(STUBBED) called");
        IPC::ResponseBuilder rb{ctx, 2, 0, 1};
        rb.Push(RESULT_SUCCESS);
@@ -783,7 +784,7 @@ private:
    }
    void GetSystemDisplayService(Kernel::HLERequestContext& ctx) {
-        LOG_WARNING(Service_VI, "(STUBBED) called");
+        NGLOG_WARNING(Service_VI, "(STUBBED) called");
        IPC::ResponseBuilder rb{ctx, 2, 0, 1};
        rb.Push(RESULT_SUCCESS);
@@ -791,7 +792,7 @@ private:
    }
    void GetManagerDisplayService(Kernel::HLERequestContext& ctx) {
-        LOG_WARNING(Service_VI, "(STUBBED) called");
+        NGLOG_WARNING(Service_VI, "(STUBBED) called");
        IPC::ResponseBuilder rb{ctx, 2, 0, 1};
        rb.Push(RESULT_SUCCESS);
@@ -799,7 +800,7 @@ private:
    }
    void GetIndirectDisplayTransactionService(Kernel::HLERequestContext& ctx) {
-        LOG_WARNING(Service_VI, "(STUBBED) called");
+        NGLOG_WARNING(Service_VI, "(STUBBED) called");
        IPC::ResponseBuilder rb{ctx, 2, 0, 1};
        rb.Push(RESULT_SUCCESS);
@@ -807,7 +808,7 @@ private:
    }
    void OpenDisplay(Kernel::HLERequestContext& ctx) {
-        LOG_WARNING(Service_VI, "(STUBBED) called");
+        NGLOG_WARNING(Service_VI, "(STUBBED) called");
        IPC::RequestParser rp{ctx};
        auto name_buf = rp.PopRaw<std::array<u8, 0x40>>();
        auto end = std::find(name_buf.begin(), name_buf.end(), '\0');
@@ -822,7 +823,7 @@ private:
    }
    void CloseDisplay(Kernel::HLERequestContext& ctx) {
-        LOG_WARNING(Service_VI, "(STUBBED) called");
+        NGLOG_WARNING(Service_VI, "(STUBBED) called");
        IPC::RequestParser rp{ctx};
        u64 display_id = rp.Pop<u64>();
@@ -831,7 +832,7 @@ private:
    }
    void GetDisplayResolution(Kernel::HLERequestContext& ctx) {
-        LOG_WARNING(Service_VI, "(STUBBED) called");
+        NGLOG_WARNING(Service_VI, "(STUBBED) called");
        IPC::RequestParser rp{ctx};
        u64 display_id = rp.Pop<u64>();
@@ -848,7 +849,7 @@ private:
    }
    void SetLayerScalingMode(Kernel::HLERequestContext& ctx) {
-        LOG_WARNING(Service_VI, "(STUBBED) called");
+        NGLOG_WARNING(Service_VI, "(STUBBED) called");
        IPC::RequestParser rp{ctx};
        u32 scaling_mode = rp.Pop<u32>();
        u64 unknown = rp.Pop<u64>();
@@ -864,11 +865,11 @@ private:
        IPC::ResponseBuilder rb = rp.MakeBuilder(4, 0, 0);
        rb.Push(RESULT_SUCCESS);
        rb.Push<u64>(1);
-        LOG_WARNING(Service_VI, "(STUBBED) called");
+        NGLOG_WARNING(Service_VI, "(STUBBED) called");
    }
    void OpenLayer(Kernel::HLERequestContext& ctx) {
-        LOG_DEBUG(Service_VI, "called");
+        NGLOG_DEBUG(Service_VI, "called");
        IPC::RequestParser rp{ctx};
        auto name_buf = rp.PopRaw<std::array<u8, 0x40>>();
        auto end = std::find(name_buf.begin(), name_buf.end(), '\0');
@@ -888,7 +889,7 @@ private:
    }
    void CreateStrayLayer(Kernel::HLERequestContext& ctx) {
-        LOG_DEBUG(Service_VI, "called");
+        NGLOG_DEBUG(Service_VI, "called");
        IPC::RequestParser rp{ctx};
        u32 flags = rp.Pop<u32>();
@@ -908,7 +909,7 @@ private:
    }
    void DestroyStrayLayer(Kernel::HLERequestContext& ctx) {
-        LOG_WARNING(Service_VI, "(STUBBED) called");
+        NGLOG_WARNING(Service_VI, "(STUBBED) called");
        IPC::RequestParser rp{ctx};
        u64 layer_id = rp.Pop<u64>();
@@ -918,7 +919,7 @@ private:
    }
    void GetDisplayVsyncEvent(Kernel::HLERequestContext& ctx) {
-        LOG_WARNING(Service_VI, "(STUBBED) called");
+        NGLOG_WARNING(Service_VI, "(STUBBED) called");
        IPC::RequestParser rp{ctx};
        u64 display_id = rp.Pop<u64>();
@@ -967,7 +968,7 @@ Module::Interface::Interface(std::shared_ptr<Module> module, const char* name,
    : ServiceFramework(name), module(std::move(module)), nv_flinger(std::move(nv_flinger)) {}
 void Module::Interface::GetDisplayService(Kernel::HLERequestContext& ctx) {
-    LOG_WARNING(Service_VI, "(STUBBED) called");
+    NGLOG_WARNING(Service_VI, "(STUBBED) called");
    IPC::ResponseBuilder rb{ctx, 2, 0, 1};
    rb.Push(RESULT_SUCCESS);
diff --git a/src/core/hw/hw.cpp b/src/core/hw/hw.cpp
index 0db604c76..f8a40390a 100644
--- a/src/core/hw/hw.cpp
+++ b/src/core/hw/hw.cpp
@@ -33,7 +33,8 @@ inline void Read(T& var, const u32 addr) {
        LCD::Read(var, addr);
        break;
    default:
-        LOG_ERROR(HW_Memory, "unknown Read%lu @ 0x%08X", sizeof(var) * 8, addr);
+        NGLOG_ERROR(HW_Memory, "Unknown Read{} @ {:#010X}", sizeof(var) * 8, addr);
+        break;
    }
 }
@@ -61,7 +62,8 @@ inline void Write(u32 addr, const T data) {
        LCD::Write(addr, data);
        break;
    default:
-        LOG_ERROR(HW_Memory, "unknown Write%lu 0x%08X @ 0x%08X", sizeof(data) * 8, (u32)data, addr);
+        NGLOG_ERROR(HW_Memory, "Unknown Write{} {:#010X} @ {:#010X}", sizeof(data) * 8, data, addr);
+        break;
    }
 }
@@ -83,12 +85,12 @@ void Update() {}
 /// Initialize hardware
 void Init() {
    LCD::Init();
-    LOG_DEBUG(HW, "initialized OK");
+    NGLOG_DEBUG(HW, "Initialized OK");
 }
 /// Shutdown hardware
 void Shutdown() {
    LCD::Shutdown();
-    LOG_DEBUG(HW, "shutdown OK");
+    NGLOG_DEBUG(HW, "Shutdown OK");
 }
 } // namespace HW
diff --git a/src/core/hw/lcd.cpp b/src/core/hw/lcd.cpp
index 690079b65..7d0046bf3 100644
--- a/src/core/hw/lcd.cpp
+++ b/src/core/hw/lcd.cpp
@@ -20,7 +20,7 @@ inline void Read(T& var, const u32 raw_addr) {
    // Reads other than u32 are untested, so I'd rather have them abort than silently fail
    if (index >= 0x400 || !std::is_same<T, u32>::value) {
-        LOG_ERROR(HW_LCD, "unknown Read%lu @ 0x%08X", sizeof(var) * 8, addr);
+        NGLOG_ERROR(HW_LCD, "Unknown Read{} @ {:#010X}", sizeof(var) * 8, addr);
        return;
    }
@@ -34,7 +34,7 @@ inline void Write(u32 addr, const T data) {
    // Writes other than u32 are untested, so I'd rather have them abort than silently fail
    if (index >= 0x400 || !std::is_same<T, u32>::value) {
-        LOG_ERROR(HW_LCD, "unknown Write%lu 0x%08X @ 0x%08X", sizeof(data) * 8, (u32)data, addr);
+        NGLOG_ERROR(HW_LCD, "Unknown Write{} {:#010X} @ {:#010X}", sizeof(data) * 8, data, addr);
        return;
    }
@@ -56,12 +56,12 @@ template void Write<u8>(u32 addr, const u8 data);
 /// Initialize hardware
 void Init() {
    memset(&g_regs, 0, sizeof(g_regs));
-    LOG_DEBUG(HW_LCD, "initialized OK");
+    NGLOG_DEBUG(HW_LCD, "Initialized OK");
 }
 /// Shutdown hardware
 void Shutdown() {
-    LOG_DEBUG(HW_LCD, "shutdown OK");
+    NGLOG_DEBUG(HW_LCD, "Shutdown OK");
 }
 } // namespace LCD
diff --git a/src/core/loader/deconstructed_rom_directory.cpp b/src/core/loader/deconstructed_rom_directory.cpp
index 8696c28bd..40a81025f 100644
--- a/src/core/loader/deconstructed_rom_directory.cpp
+++ b/src/core/loader/deconstructed_rom_directory.cpp
@@ -132,7 +132,7 @@ ResultStatus AppLoader_DeconstructedRomDirectory::Load(
        const VAddr load_addr = next_load_addr;
        next_load_addr = AppLoader_NSO::LoadModule(path, load_addr);
        if (next_load_addr) {
-            LOG_DEBUG(Loader, "loaded module %s @ 0x%" PRIx64, module, load_addr);
+            NGLOG_DEBUG(Loader, "loaded module {} @ {:#X}", module, load_addr);
        } else {
            next_load_addr = load_addr;
        }
@@ -163,7 +163,7 @@ ResultStatus AppLoader_DeconstructedRomDirectory::ReadRomFS(
    std::shared_ptr<FileUtil::IOFile>& romfs_file, u64& offset, u64& size) {
    if (filepath_romfs.empty()) {
-        LOG_DEBUG(Loader, "No RomFS available");
+        NGLOG_DEBUG(Loader, "No RomFS available");
        return ResultStatus::ErrorNotUsed;
    }
@@ -176,8 +176,8 @@ ResultStatus AppLoader_DeconstructedRomDirectory::ReadRomFS(
    offset = 0;
    size = romfs_file->GetSize();
-    LOG_DEBUG(Loader, "RomFS offset:           0x%016" PRIX64, offset);
+    NGLOG_DEBUG(Loader, "RomFS offset:           {:#018X}", offset);
-    LOG_DEBUG(Loader, "RomFS size:             0x%016" PRIX64, size);
+    NGLOG_DEBUG(Loader, "RomFS size:             {:#018X}", size);
    // Reset read pointer
    file.Seek(0, SEEK_SET);
diff --git a/src/core/loader/elf.cpp b/src/core/loader/elf.cpp
index e9f462196..e42d3a870 100644
--- a/src/core/loader/elf.cpp
+++ b/src/core/loader/elf.cpp
@@ -273,18 +273,18 @@ const char* ElfReader::GetSectionName(int section) const {
 }
 SharedPtr<CodeSet> ElfReader::LoadInto(u32 vaddr) {
-    LOG_DEBUG(Loader, "String section: %i", header->e_shstrndx);
+    NGLOG_DEBUG(Loader, "String section: {}", header->e_shstrndx);
    // Should we relocate?
    relocate = (header->e_type != ET_EXEC);
    if (relocate) {
-        LOG_DEBUG(Loader, "Relocatable module");
+        NGLOG_DEBUG(Loader, "Relocatable module");
        entryPoint += vaddr;
    } else {
-        LOG_DEBUG(Loader, "Prerelocated executable");
+        NGLOG_DEBUG(Loader, "Prerelocated executable");
    }
-    LOG_DEBUG(Loader, "%i segments:", header->e_phnum);
+    NGLOG_DEBUG(Loader, "{} segments:", header->e_phnum);
    // First pass : Get the bits into RAM
    u32 base_addr = relocate ? vaddr : 0;
@@ -304,8 +304,8 @@ SharedPtr<CodeSet> ElfReader::LoadInto(u32 vaddr) {
    for (unsigned int i = 0; i < header->e_phnum; ++i) {
        Elf32_Phdr* p = &segments[i];
-        LOG_DEBUG(Loader, "Type: %i Vaddr: %08X Filesz: %8X Memsz: %8X ", p->p_type, p->p_vaddr,
+        NGLOG_DEBUG(Loader, "Type: {} Vaddr: {:08X} Filesz: {:08X} Memsz: {:08X} ", p->p_type,
-                  p->p_filesz, p->p_memsz);
+                    p->p_vaddr, p->p_filesz, p->p_memsz);
        if (p->p_type == PT_LOAD) {
            CodeSet::Segment* codeset_segment;
@@ -317,16 +317,16 @@ SharedPtr<CodeSet> ElfReader::LoadInto(u32 vaddr) {
            } else if (permission_flags == (PF_R | PF_W)) {
                codeset_segment = &codeset->data;
            } else {
-                LOG_ERROR(Loader, "Unexpected ELF PT_LOAD segment id %u with flags %X", i,
+                NGLOG_ERROR(Loader, "Unexpected ELF PT_LOAD segment id {} with flags {:X}", i,
-                          p->p_flags);
+                            p->p_flags);
                continue;
            }
            if (codeset_segment->size != 0) {
-                LOG_ERROR(Loader,
+                NGLOG_ERROR(Loader,
-                          "ELF has more than one segment of the same type. Skipping extra "
+                            "ELF has more than one segment of the same type. Skipping extra "
-                          "segment (id %i)",
+                            "segment (id {})",
-                          i);
+                            i);
                continue;
            }
@@ -345,7 +345,7 @@ SharedPtr<CodeSet> ElfReader::LoadInto(u32 vaddr) {
    codeset->entrypoint = base_addr + header->e_entry;
    codeset->memory = std::make_shared<std::vector<u8>>(std::move(program_image));
-    LOG_DEBUG(Loader, "Done loading.");
+    NGLOG_DEBUG(Loader, "Done loading.");
    return codeset;
 }
diff --git a/src/core/loader/linker.cpp b/src/core/loader/linker.cpp
index 69198e3e3..c7be5f265 100644
--- a/src/core/loader/linker.cpp
+++ b/src/core/loader/linker.cpp
@@ -84,7 +84,7 @@ void Linker::WriteRelocations(std::vector<u8>& program_image, const std::vector<
            }
            break;
        default:
-            LOG_CRITICAL(Loader, "Unknown relocation type: %d", static_cast<int>(rela.type));
+            NGLOG_CRITICAL(Loader, "Unknown relocation type: {}", static_cast<int>(rela.type));
            break;
        }
    }
@@ -141,7 +141,7 @@ void Linker::ResolveImports() {
        if (search != exports.end()) {
            Memory::Write64(import.second.ea, search->second + import.second.addend);
        } else {
-            LOG_ERROR(Loader, "Unresolved import: %s", import.first.c_str());
+            NGLOG_ERROR(Loader, "Unresolved import: {}", import.first);
        }
    }
 }
diff --git a/src/core/loader/loader.cpp b/src/core/loader/loader.cpp
index 2ec08506d..6a4fd38cb 100644
--- a/src/core/loader/loader.cpp
+++ b/src/core/loader/loader.cpp
@@ -41,7 +41,7 @@ FileType IdentifyFile(FileUtil::IOFile& file, const std::string& filepath) {
 FileType IdentifyFile(const std::string& file_name) {
    FileUtil::IOFile file(file_name, "rb");
    if (!file.IsOpen()) {
-        LOG_ERROR(Loader, "Failed to load file %s", file_name.c_str());
+        NGLOG_ERROR(Loader, "Failed to load file {}", file_name);
        return FileType::Unknown;
    }
@@ -116,7 +116,7 @@ static std::unique_ptr<AppLoader> GetFileLoader(FileUtil::IOFile&& file, FileTyp
 std::unique_ptr<AppLoader> GetLoader(const std::string& filename) {
    FileUtil::IOFile file(filename, "rb");
    if (!file.IsOpen()) {
-        LOG_ERROR(Loader, "Failed to load file %s", filename.c_str());
+        NGLOG_ERROR(Loader, "Failed to load file {}", filename);
        return nullptr;
    }
@@ -127,12 +127,12 @@ std::unique_ptr<AppLoader> GetLoader(const std::string& filename) {
    FileType filename_type = GuessFromExtension(filename_extension);
    if (type != filename_type) {
-        LOG_WARNING(Loader, "File %s has a different type than its extension.", filename.c_str());
+        NGLOG_WARNING(Loader, "File {} has a different type than its extension.", filename);
        if (FileType::Unknown == type)
            type = filename_type;
    }
-    LOG_DEBUG(Loader, "Loading file %s as %s...", filename.c_str(), GetFileTypeString(type));
+    NGLOG_DEBUG(Loader, "Loading file {} as {}...", filename, GetFileTypeString(type));
    return GetFileLoader(std::move(file), type, filename_filename, filename);
 }
diff --git a/src/core/loader/nro.cpp b/src/core/loader/nro.cpp
index b5133e4d6..3853cfa1a 100644
--- a/src/core/loader/nro.cpp
+++ b/src/core/loader/nro.cpp
@@ -137,7 +137,7 @@ ResultStatus AppLoader_NRO::Load(Kernel::SharedPtr<Kernel::Process>& process) {
    process->address_mappings = default_address_mappings;
    process->resource_limit =
        Kernel::ResourceLimit::GetForCategory(Kernel::ResourceLimitCategory::APPLICATION);
-    process->Run(base_addr, 48, Memory::DEFAULT_STACK_SIZE);
+    process->Run(base_addr, THREADPRIO_DEFAULT, Memory::DEFAULT_STACK_SIZE);
    is_loaded = true;
    return ResultStatus::Success;
diff --git a/src/core/loader/nso.cpp b/src/core/loader/nso.cpp
index 3bc10ed0d..1842bae20 100644
--- a/src/core/loader/nso.cpp
+++ b/src/core/loader/nso.cpp
@@ -73,7 +73,7 @@ static std::vector<u8> ReadSegment(FileUtil::IOFile& file, const NsoSegmentHeade
    file.Seek(header.offset, SEEK_SET);
    if (compressed_size != file.ReadBytes(compressed_data.data(), compressed_size)) {
-        LOG_CRITICAL(Loader, "Failed to read %d NSO LZ4 compressed bytes", compressed_size);
+        NGLOG_CRITICAL(Loader, "Failed to read {} NSO LZ4 compressed bytes", compressed_size);
        return {};
    }
@@ -158,14 +158,13 @@ ResultStatus AppLoader_NSO::Load(Kernel::SharedPtr<Kernel::Process>& process) {
    // Load module
    LoadModule(filepath, Memory::PROCESS_IMAGE_VADDR);
-    LOG_DEBUG(Loader, "loaded module %s @ 0x%" PRIx64, filepath.c_str(),
+    NGLOG_DEBUG(Loader, "loaded module {} @ {:#X}", filepath, Memory::PROCESS_IMAGE_VADDR);
-              Memory::PROCESS_IMAGE_VADDR);
    process->svc_access_mask.set();
    process->address_mappings = default_address_mappings;
    process->resource_limit =
        Kernel::ResourceLimit::GetForCategory(Kernel::ResourceLimitCategory::APPLICATION);
-    process->Run(Memory::PROCESS_IMAGE_VADDR, 48, Memory::DEFAULT_STACK_SIZE);
+    process->Run(Memory::PROCESS_IMAGE_VADDR, THREADPRIO_DEFAULT, Memory::DEFAULT_STACK_SIZE);
    is_loaded = true;
    return ResultStatus::Success;
diff --git a/src/core/memory.cpp b/src/core/memory.cpp
index 291bf066f..2afa0916d 100644
--- a/src/core/memory.cpp
+++ b/src/core/memory.cpp
@@ -39,8 +39,8 @@ PageTable* GetCurrentPageTable() {
 }
 static void MapPages(PageTable& page_table, VAddr base, u64 size, u8* memory, PageType type) {
-    LOG_DEBUG(HW_Memory, "Mapping %p onto %016" PRIX64 "-%016" PRIX64, memory, base * PAGE_SIZE,
+    NGLOG_DEBUG(HW_Memory, "Mapping {} onto {:016X}-{:016X}", fmt::ptr(memory), base * PAGE_SIZE,
-              (base + size) * PAGE_SIZE);
+                (base + size) * PAGE_SIZE);
    RasterizerFlushVirtualRegion(base << PAGE_BITS, size * PAGE_SIZE,
                                 FlushMode::FlushAndInvalidate);
@@ -169,10 +169,10 @@ T Read(const VAddr vaddr) {
    PageType type = current_page_table->attributes[vaddr >> PAGE_BITS];
    switch (type) {
    case PageType::Unmapped:
-        LOG_ERROR(HW_Memory, "unmapped Read%lu @ 0x%08X", sizeof(T) * 8, vaddr);
+        NGLOG_ERROR(HW_Memory, "Unmapped Read{} @ {:#010X}", sizeof(T) * 8, vaddr);
        return 0;
    case PageType::Memory:
-        ASSERT_MSG(false, "Mapped memory page without a pointer @ %08X", vaddr);
+        ASSERT_MSG(false, "Mapped memory page without a pointer @ %016" PRIX64, vaddr);
        break;
    case PageType::RasterizerCachedMemory: {
        RasterizerFlushVirtualRegion(vaddr, sizeof(T), FlushMode::Flush);
@@ -201,11 +201,11 @@ void Write(const VAddr vaddr, const T data) {
    PageType type = current_page_table->attributes[vaddr >> PAGE_BITS];
    switch (type) {
    case PageType::Unmapped:
-        LOG_ERROR(HW_Memory, "unmapped Write%lu 0x%08X @ 0x%08X", sizeof(data) * 8, (u32)data,
+        NGLOG_ERROR(HW_Memory, "Unmapped Write{} {:#010X} @ {:#018X}", sizeof(data) * 8, (u32)data,
-                  vaddr);
+                    vaddr);
        return;
    case PageType::Memory:
-        ASSERT_MSG(false, "Mapped memory page without a pointer @ %08X", vaddr);
+        ASSERT_MSG(false, "Mapped memory page without a pointer @ %016" PRIX64, vaddr);
        break;
    case PageType::RasterizerCachedMemory: {
        RasterizerFlushVirtualRegion(vaddr, sizeof(T), FlushMode::Invalidate);
@@ -251,7 +251,7 @@ u8* GetPointer(const VAddr vaddr) {
        return GetPointerFromVMA(vaddr);
    }
-    LOG_ERROR(HW_Memory, "unknown GetPointer @ 0x%08x", vaddr);
+    NGLOG_ERROR(HW_Memory, "Unknown GetPointer @ {:#018X}", vaddr);
    return nullptr;
 }
@@ -288,13 +288,12 @@ u8* GetPhysicalPointer(PAddr address) {
        });
    if (area == std::end(memory_areas)) {
-        LOG_ERROR(HW_Memory, "unknown GetPhysicalPointer @ 0x%016" PRIX64, address);
+        NGLOG_ERROR(HW_Memory, "Unknown GetPhysicalPointer @ {:#018X}", address);
        return nullptr;
    }
    if (area->paddr_base == IO_AREA_PADDR) {
-        LOG_ERROR(HW_Memory, "MMIO mappings are not supported yet. phys_addr=0x%016" PRIX64,
+        NGLOG_ERROR(HW_Memory, "MMIO mappings are not supported yet. phys_addr={:018X}", address);
-                  address);
        return nullptr;
    }
@@ -325,15 +324,29 @@ u8* GetPhysicalPointer(PAddr address) {
    return target_pointer;
 }
-void RasterizerMarkRegionCached(VAddr start, u64 size, bool cached) {
+void RasterizerMarkRegionCached(Tegra::GPUVAddr gpu_addr, u64 size, bool cached) {
-    if (start == 0) {
+    if (gpu_addr == 0) {
        return;
    }
-    u64 num_pages = ((start + size - 1) >> PAGE_BITS) - (start >> PAGE_BITS) + 1;
+    // Iterate over a contiguous CPU address space, which corresponds to the specified GPU address
-    VAddr vaddr = start;
+    // space, marking the region as un/cached. The region is marked un/cached at a granularity of
+    // CPU pages, hence why we iterate on a CPU page basis (note: GPU page size is different). This
+    // assumes the specified GPU address region is contiguous as well.
+    u64 num_pages = ((gpu_addr + size - 1) >> PAGE_BITS) - (gpu_addr >> PAGE_BITS) + 1;
+    for (unsigned i = 0; i < num_pages; ++i, gpu_addr += PAGE_SIZE) {
+        boost::optional<VAddr> maybe_vaddr =
+            Core::System::GetInstance().GPU().memory_manager->GpuToCpuAddress(gpu_addr);
+        // The GPU <-> CPU virtual memory mapping is not 1:1
+        if (!maybe_vaddr) {
+            NGLOG_ERROR(HW_Memory,
+                        "Trying to flush a cached region to an invalid physical address {:016X}",
+                        gpu_addr);
+            continue;
+        }
+        VAddr vaddr = *maybe_vaddr;
-    for (unsigned i = 0; i < num_pages; ++i, vaddr += PAGE_SIZE) {
        PageType& page_type = current_page_table->attributes[vaddr >> PAGE_BITS];
        if (cached) {
@@ -347,6 +360,10 @@ void RasterizerMarkRegionCached(VAddr start, u64 size, bool cached) {
                page_type = PageType::RasterizerCachedMemory;
                current_page_table->pointers[vaddr >> PAGE_BITS] = nullptr;
                break;
+            case PageType::RasterizerCachedMemory:
+                // There can be more than one GPU region mapped per CPU region, so it's common that
+                // this area is already marked as cached.
+                break;
            default:
                UNREACHABLE();
            }
@@ -357,6 +374,10 @@ void RasterizerMarkRegionCached(VAddr start, u64 size, bool cached) {
                // It is not necessary for a process to have this region mapped into its address
                // space, for example, a system module need not have a VRAM mapping.
                break;
+            case PageType::Memory:
+                // There can be more than one GPU region mapped per CPU region, so it's common that
+                // this area is already unmarked as cached.
+                break;
            case PageType::RasterizerCachedMemory: {
                u8* pointer = GetPointerFromVMA(vaddr & ~PAGE_MASK);
                if (pointer == nullptr) {
@@ -394,19 +415,29 @@ void RasterizerFlushVirtualRegion(VAddr start, u64 size, FlushMode mode) {
        VAddr overlap_start = std::max(start, region_start);
        VAddr overlap_end = std::min(end, region_end);
+        std::vector<Tegra::GPUVAddr> gpu_addresses =
+            Core::System::GetInstance().GPU().memory_manager->CpuToGpuAddress(overlap_start);
+        if (gpu_addresses.empty()) {
+            return;
+        }
        u64 overlap_size = overlap_end - overlap_start;
-        auto* rasterizer = VideoCore::g_renderer->Rasterizer();
+        for (const auto& gpu_address : gpu_addresses) {
-        switch (mode) {
+            auto* rasterizer = VideoCore::g_renderer->Rasterizer();
-        case FlushMode::Flush:
+            switch (mode) {
-            rasterizer->FlushRegion(overlap_start, overlap_size);
+            case FlushMode::Flush:
-            break;
+                rasterizer->FlushRegion(gpu_address, overlap_size);
-        case FlushMode::Invalidate:
+                break;
-            rasterizer->InvalidateRegion(overlap_start, overlap_size);
+            case FlushMode::Invalidate:
-            break;
+                rasterizer->InvalidateRegion(gpu_address, overlap_size);
-        case FlushMode::FlushAndInvalidate:
+                break;
-            rasterizer->FlushAndInvalidateRegion(overlap_start, overlap_size);
+            case FlushMode::FlushAndInvalidate:
-            break;
+                rasterizer->FlushAndInvalidateRegion(gpu_address, overlap_size);
+                break;
+            }
        }
    };
@@ -445,8 +476,9 @@ void ReadBlock(const Kernel::Process& process, const VAddr src_addr, void* dest_
        switch (page_table.attributes[page_index]) {
        case PageType::Unmapped: {
-            LOG_ERROR(HW_Memory, "unmapped ReadBlock @ 0x%08X (start address = 0x%08X, size = %zu)",
+            NGLOG_ERROR(HW_Memory,
-                      current_vaddr, src_addr, size);
+                        "Unmapped ReadBlock @ {:#018X} (start address = {:#018X}, size = {})",
+                        current_vaddr, src_addr, size);
            std::memset(dest_buffer, 0, copy_amount);
            break;
        }
@@ -508,9 +540,9 @@ void WriteBlock(const Kernel::Process& process, const VAddr dest_addr, const voi
        switch (page_table.attributes[page_index]) {
        case PageType::Unmapped: {
-            LOG_ERROR(HW_Memory,
+            NGLOG_ERROR(HW_Memory,
-                      "unmapped WriteBlock @ 0x%08X (start address = 0x%08X, size = %zu)",
+                        "Unmapped WriteBlock @ {:#018X} (start address = {:#018X}, size = {})",
-                      current_vaddr, dest_addr, size);
+                        current_vaddr, dest_addr, size);
            break;
        }
        case PageType::Memory: {
@@ -556,8 +588,9 @@ void ZeroBlock(const Kernel::Process& process, const VAddr dest_addr, const size
        switch (page_table.attributes[page_index]) {
        case PageType::Unmapped: {
-            LOG_ERROR(HW_Memory, "unmapped ZeroBlock @ 0x%08X (start address = 0x%08X, size = %zu)",
+            NGLOG_ERROR(HW_Memory,
-                      current_vaddr, dest_addr, size);
+                        "Unmapped ZeroBlock @ {:#018X} (start address = {#:018X}, size = {})",
+                        current_vaddr, dest_addr, size);
            break;
        }
        case PageType::Memory: {
@@ -596,8 +629,9 @@ void CopyBlock(const Kernel::Process& process, VAddr dest_addr, VAddr src_addr,
        switch (page_table.attributes[page_index]) {
        case PageType::Unmapped: {
-            LOG_ERROR(HW_Memory, "unmapped CopyBlock @ 0x%08X (start address = 0x%08X, size = %zu)",
+            NGLOG_ERROR(HW_Memory,
-                      current_vaddr, src_addr, size);
+                        "Unmapped CopyBlock @ {:#018X} (start address = {:#018X}, size = {})",
+                        current_vaddr, src_addr, size);
            ZeroBlock(process, dest_addr, copy_amount);
            break;
        }
@@ -625,6 +659,10 @@ void CopyBlock(const Kernel::Process& process, VAddr dest_addr, VAddr src_addr,
    }
 }
+void CopyBlock(VAddr dest_addr, VAddr src_addr, size_t size) {
+    CopyBlock(*Core::CurrentProcess(), dest_addr, src_addr, size);
+}
 boost::optional<PAddr> TryVirtualToPhysicalAddress(const VAddr addr) {
    if (addr == 0) {
        return 0;
@@ -646,7 +684,7 @@ boost::optional<PAddr> TryVirtualToPhysicalAddress(const VAddr addr) {
 PAddr VirtualToPhysicalAddress(const VAddr addr) {
    auto paddr = TryVirtualToPhysicalAddress(addr);
    if (!paddr) {
-        LOG_ERROR(HW_Memory, "Unknown virtual address @ 0x%016" PRIX64, addr);
+        NGLOG_ERROR(HW_Memory, "Unknown virtual address @ {:#018X}", addr);
        // To help with debugging, set bit on address so that it's obviously invalid.
        return addr | 0x80000000;
    }
diff --git a/src/core/memory.h b/src/core/memory.h
index e9b8ca873..3f56a2c6a 100644
--- a/src/core/memory.h
+++ b/src/core/memory.h
@@ -14,6 +14,7 @@
 #include <boost/optional.hpp>
 #include "common/common_types.h"
 #include "core/memory_hook.h"
+#include "video_core/memory_manager.h"
 namespace Kernel {
 class Process;
@@ -258,7 +259,7 @@ enum class FlushMode {
 /**
 * Mark each page touching the region as cached.
 */
-void RasterizerMarkRegionCached(VAddr start, u64 size, bool cached);
+void RasterizerMarkRegionCached(Tegra::GPUVAddr start, u64 size, bool cached);
 /**
 * Flushes and invalidates any externally cached rasterizer resources touching the given virtual
diff --git a/src/video_core/command_processor.cpp b/src/video_core/command_processor.cpp
index d4cdb4ab2..2eaece298 100644
--- a/src/video_core/command_processor.cpp
+++ b/src/video_core/command_processor.cpp
@@ -24,41 +24,18 @@ namespace Tegra {
 enum class BufferMethods {
    BindObject = 0,
-    SetGraphMacroCode = 0x45,
+    CountBufferMethods = 0x40,
-    SetGraphMacroCodeArg = 0x46,
-    SetGraphMacroEntry = 0x47,
-    CountBufferMethods = 0x100,
 };
 void GPU::WriteReg(u32 method, u32 subchannel, u32 value, u32 remaining_params) {
-    LOG_WARNING(HW_GPU, "Processing method %08X on subchannel %u value %08X remaining params %u",
+    NGLOG_WARNING(HW_GPU,
-                method, subchannel, value, remaining_params);
+                  "Processing method {:08X} on subchannel {} value "
+                  "{:08X} remaining params {}",
-    if (method == static_cast<u32>(BufferMethods::SetGraphMacroEntry)) {
+                  method, subchannel, value, remaining_params);
-        // Prepare to upload a new macro, reset the upload counter.
-        LOG_DEBUG(HW_GPU, "Uploading GPU macro %08X", value);
-        current_macro_entry = value;
-        current_macro_code.clear();
-        return;
-    }
-    if (method == static_cast<u32>(BufferMethods::SetGraphMacroCodeArg)) {
-        // Append a new code word to the current macro.
-        current_macro_code.push_back(value);
-        // There are no more params remaining, submit the code to the 3D engine.
-        if (remaining_params == 0) {
-            maxwell_3d->SubmitMacroCode(current_macro_entry, std::move(current_macro_code));
-            current_macro_entry = InvalidGraphMacroEntry;
-            current_macro_code.clear();
-        }
-        return;
-    }
    if (method == static_cast<u32>(BufferMethods::BindObject)) {
        // Bind the current subchannel to the desired engine id.
-        LOG_DEBUG(HW_GPU, "Binding subchannel %u to engine %u", subchannel, value);
+        NGLOG_DEBUG(HW_GPU, "Binding subchannel {} to engine {}", subchannel, value);
        ASSERT(bound_engines.find(subchannel) == bound_engines.end());
        bound_engines[subchannel] = static_cast<EngineID>(value);
        return;
@@ -66,7 +43,7 @@ void GPU::WriteReg(u32 method, u32 subchannel, u32 value, u32 remaining_params)
    if (method < static_cast<u32>(BufferMethods::CountBufferMethods)) {
        // TODO(Subv): Research and implement these methods.
-        LOG_ERROR(HW_GPU, "Special buffer methods other than Bind are not implemented");
+        NGLOG_ERROR(HW_GPU, "Special buffer methods other than Bind are not implemented");
        return;
    }
@@ -90,11 +67,9 @@ void GPU::WriteReg(u32 method, u32 subchannel, u32 value, u32 remaining_params)
 }
 void GPU::ProcessCommandList(GPUVAddr address, u32 size) {
-    // TODO(Subv): PhysicalToVirtualAddress is a misnomer, it converts a GPU VAddr into an
+    const boost::optional<VAddr> head_address = memory_manager->GpuToCpuAddress(address);
-    // application VAddr.
+    VAddr current_addr = *head_address;
-    const VAddr head_address = memory_manager->PhysicalToVirtualAddress(address);
+    while (current_addr < *head_address + size * sizeof(CommandHeader)) {
-    VAddr current_addr = head_address;
-    while (current_addr < head_address + size * sizeof(CommandHeader)) {
        const CommandHeader header = {Memory::Read32(current_addr)};
        current_addr += sizeof(u32);
diff --git a/src/video_core/engines/fermi_2d.cpp b/src/video_core/engines/fermi_2d.cpp
index 7aab163dc..9019f2504 100644
--- a/src/video_core/engines/fermi_2d.cpp
+++ b/src/video_core/engines/fermi_2d.cpp
@@ -2,12 +2,71 @@
 // Licensed under GPLv2 or any later version
 // Refer to the license.txt file included.
+#include "core/memory.h"
 #include "video_core/engines/fermi_2d.h"
+#include "video_core/textures/decoders.h"
 namespace Tegra {
 namespace Engines {
-void Fermi2D::WriteReg(u32 method, u32 value) {}
+Fermi2D::Fermi2D(MemoryManager& memory_manager) : memory_manager(memory_manager) {}
+void Fermi2D::WriteReg(u32 method, u32 value) {
+    ASSERT_MSG(method < Regs::NUM_REGS,
+               "Invalid Fermi2D register, increase the size of the Regs structure");
+    regs.reg_array[method] = value;
+    switch (method) {
+    case FERMI2D_REG_INDEX(trigger): {
+        HandleSurfaceCopy();
+        break;
+    }
+    }
+}
+void Fermi2D::HandleSurfaceCopy() {
+    NGLOG_WARNING(HW_GPU, "Requested a surface copy with operation {}",
+                  static_cast<u32>(regs.operation));
+    const GPUVAddr source = regs.src.Address();
+    const GPUVAddr dest = regs.dst.Address();
+    // TODO(Subv): Only same-format and same-size copies are allowed for now.
+    ASSERT(regs.src.format == regs.dst.format);
+    ASSERT(regs.src.width * regs.src.height == regs.dst.width * regs.dst.height);
+    // TODO(Subv): Only raw copies are implemented.
+    ASSERT(regs.operation == Regs::Operation::SrcCopy);
+    const VAddr source_cpu = *memory_manager.GpuToCpuAddress(source);
+    const VAddr dest_cpu = *memory_manager.GpuToCpuAddress(dest);
+    u32 src_bytes_per_pixel = RenderTargetBytesPerPixel(regs.src.format);
+    u32 dst_bytes_per_pixel = RenderTargetBytesPerPixel(regs.dst.format);
+    if (regs.src.linear == regs.dst.linear) {
+        // If the input layout and the output layout are the same, just perform a raw copy.
+        Memory::CopyBlock(dest_cpu, source_cpu,
+                          src_bytes_per_pixel * regs.dst.width * regs.dst.height);
+        return;
+    }
+    u8* src_buffer = Memory::GetPointer(source_cpu);
+    u8* dst_buffer = Memory::GetPointer(dest_cpu);
+    if (!regs.src.linear && regs.dst.linear) {
+        // If the input is tiled and the output is linear, deswizzle the input and copy it over.
+        Texture::CopySwizzledData(regs.src.width, regs.src.height, src_bytes_per_pixel,
+                                  dst_bytes_per_pixel, src_buffer, dst_buffer, true,
+                                  regs.src.block_height);
+    } else {
+        // If the input is linear and the output is tiled, swizzle the input and copy it over.
+        Texture::CopySwizzledData(regs.src.width, regs.src.height, src_bytes_per_pixel,
+                                  dst_bytes_per_pixel, dst_buffer, src_buffer, false,
+                                  regs.dst.block_height);
+    }
+}
 } // namespace Engines
 } // namespace Tegra
diff --git a/src/video_core/engines/fermi_2d.h b/src/video_core/engines/fermi_2d.h
index 8967ddede..0c5b413cc 100644
--- a/src/video_core/engines/fermi_2d.h
+++ b/src/video_core/engines/fermi_2d.h
@@ -4,19 +4,106 @@
 #pragma once
+#include <array>
+#include "common/assert.h"
+#include "common/bit_field.h"
+#include "common/common_funcs.h"
 #include "common/common_types.h"
+#include "video_core/gpu.h"
+#include "video_core/memory_manager.h"
 namespace Tegra {
 namespace Engines {
+#define FERMI2D_REG_INDEX(field_name)                                                              \
+    (offsetof(Tegra::Engines::Fermi2D::Regs, field_name) / sizeof(u32))
 class Fermi2D final {
 public:
-    Fermi2D() = default;
+    explicit Fermi2D(MemoryManager& memory_manager);
    ~Fermi2D() = default;
    /// Write the value to the register identified by method.
    void WriteReg(u32 method, u32 value);
+    struct Regs {
+        static constexpr size_t NUM_REGS = 0x258;
+        struct Surface {
+            RenderTargetFormat format;
+            BitField<0, 1, u32> linear;
+            union {
+                BitField<0, 4, u32> block_depth;
+                BitField<4, 4, u32> block_height;
+                BitField<8, 4, u32> block_width;
+            };
+            u32 depth;
+            u32 layer;
+            u32 pitch;
+            u32 width;
+            u32 height;
+            u32 address_high;
+            u32 address_low;
+            GPUVAddr Address() const {
+                return static_cast<GPUVAddr>((static_cast<GPUVAddr>(address_high) << 32) |
+                                             address_low);
+            }
+        };
+        static_assert(sizeof(Surface) == 0x28, "Surface has incorrect size");
+        enum class Operation : u32 {
+            SrcCopyAnd = 0,
+            ROPAnd = 1,
+            Blend = 2,
+            SrcCopy = 3,
+            ROP = 4,
+            SrcCopyPremult = 5,
+            BlendPremult = 6,
+        };
+        union {
+            struct {
+                INSERT_PADDING_WORDS(0x80);
+                Surface dst;
+                INSERT_PADDING_WORDS(2);
+                Surface src;
+                INSERT_PADDING_WORDS(0x15);
+                Operation operation;
+                INSERT_PADDING_WORDS(0x9);
+                // TODO(Subv): This is only a guess.
+                u32 trigger;
+                INSERT_PADDING_WORDS(0x1A3);
+            };
+            std::array<u32, NUM_REGS> reg_array;
+        };
+    } regs{};
+    MemoryManager& memory_manager;
+private:
+    /// Performs the copy from the source surface to the destination surface as configured in the
+    /// registers.
+    void HandleSurfaceCopy();
 };
+#define ASSERT_REG_POSITION(field_name, position)                                                  \
+    static_assert(offsetof(Fermi2D::Regs, field_name) == position * 4,                             \
+                  "Field " #field_name " has invalid position")
+ASSERT_REG_POSITION(dst, 0x80);
+ASSERT_REG_POSITION(src, 0x8C);
+ASSERT_REG_POSITION(operation, 0xAB);
+ASSERT_REG_POSITION(trigger, 0xB5);
+#undef ASSERT_REG_POSITION
 } // namespace Engines
 } // namespace Tegra
diff --git a/src/video_core/engines/maxwell_3d.cpp b/src/video_core/engines/maxwell_3d.cpp
index 2a3ff234a..4306b894f 100644
--- a/src/video_core/engines/maxwell_3d.cpp
+++ b/src/video_core/engines/maxwell_3d.cpp
@@ -22,10 +22,6 @@ constexpr u32 MacroRegistersStart = 0xE00;
 Maxwell3D::Maxwell3D(MemoryManager& memory_manager)
    : memory_manager(memory_manager), macro_interpreter(*this) {}
-void Maxwell3D::SubmitMacroCode(u32 entry, std::vector<u32> code) {
-    uploaded_macros[entry * 2 + MacroRegistersStart] = std::move(code);
-}
 void Maxwell3D::CallMacroMethod(u32 method, std::vector<u32> parameters) {
    auto macro_code = uploaded_macros.find(method);
    // The requested macro must have been uploaded already.
@@ -37,9 +33,6 @@ void Maxwell3D::CallMacroMethod(u32 method, std::vector<u32> parameters) {
 }
 void Maxwell3D::WriteReg(u32 method, u32 value, u32 remaining_params) {
-    ASSERT_MSG(method < Regs::NUM_REGS,
-               "Invalid Maxwell3D register, increase the size of the Regs structure");
    auto debug_context = Core::System::GetInstance().GetGPUDebugContext();
    // It is an error to write to a register other than the current macro's ARG register before it
@@ -68,6 +61,9 @@ void Maxwell3D::WriteReg(u32 method, u32 value, u32 remaining_params) {
        return;
    }
+    ASSERT_MSG(method < Regs::NUM_REGS,
+               "Invalid Maxwell3D register, increase the size of the Regs structure");
    if (debug_context) {
        debug_context->OnEvent(Tegra::DebugContext::Event::MaxwellCommandLoaded, nullptr);
    }
@@ -75,6 +71,10 @@ void Maxwell3D::WriteReg(u32 method, u32 value, u32 remaining_params) {
    regs.reg_array[method] = value;
    switch (method) {
+    case MAXWELL3D_REG_INDEX(macros.data): {
+        ProcessMacroUpload(value);
+        break;
+    }
    case MAXWELL3D_REG_INDEX(code_address.code_address_high):
    case MAXWELL3D_REG_INDEX(code_address.code_address_low): {
        // Note: For some reason games (like Puyo Puyo Tetris) seem to write 0 to the CODE_ADDRESS
@@ -141,17 +141,48 @@ void Maxwell3D::WriteReg(u32 method, u32 value, u32 remaining_params) {
    }
 }
+void Maxwell3D::ProcessMacroUpload(u32 data) {
+    // Store the uploaded macro code to interpret them when they're called.
+    auto& macro = uploaded_macros[regs.macros.entry * 2 + MacroRegistersStart];
+    macro.push_back(data);
+}
 void Maxwell3D::ProcessQueryGet() {
    GPUVAddr sequence_address = regs.query.QueryAddress();
    // Since the sequence address is given as a GPU VAddr, we have to convert it to an application
    // VAddr before writing.
-    VAddr address = memory_manager.PhysicalToVirtualAddress(sequence_address);
+    boost::optional<VAddr> address = memory_manager.GpuToCpuAddress(sequence_address);
+    // TODO(Subv): Support the other query units.
+    ASSERT_MSG(regs.query.query_get.unit == Regs::QueryUnit::Crop,
+               "Units other than CROP are unimplemented");
+    ASSERT_MSG(regs.query.query_get.short_query,
+               "Writing the entire query result structure is unimplemented");
+    u32 value = Memory::Read32(*address);
+    u32 result = 0;
+    // TODO(Subv): Support the other query variables
+    switch (regs.query.query_get.select) {
+    case Regs::QuerySelect::Zero:
+        result = 0;
+        break;
+    default:
+        UNIMPLEMENTED_MSG("Unimplemented query select type %u",
+                          static_cast<u32>(regs.query.query_get.select.Value()));
+    }
+    // TODO(Subv): Research and implement how query sync conditions work.
    switch (regs.query.query_get.mode) {
-    case Regs::QueryMode::Write: {
+    case Regs::QueryMode::Write:
+    case Regs::QueryMode::Write2: {
        // Write the current query sequence to the sequence address.
        u32 sequence = regs.query.query_sequence;
-        Memory::Write32(address, sequence);
+        Memory::Write32(*address, sequence);
+        // TODO(Subv): Write the proper query response structure to the address when not using short
+        // mode.
        break;
    }
    default:
@@ -161,8 +192,8 @@ void Maxwell3D::ProcessQueryGet() {
 }
 void Maxwell3D::DrawArrays() {
-    LOG_DEBUG(HW_GPU, "called, topology=%d, count=%d", regs.draw.topology.Value(),
+    NGLOG_DEBUG(HW_GPU, "called, topology={}, count={}",
-              regs.vertex_buffer.count);
+                static_cast<u32>(regs.draw.topology.Value()), regs.vertex_buffer.count);
    ASSERT_MSG(!(regs.index_array.count && regs.vertex_buffer.count), "Both indexed and direct?");
    auto debug_context = Core::System::GetInstance().GetGPUDebugContext();
@@ -200,10 +231,10 @@ void Maxwell3D::ProcessCBData(u32 value) {
    // Don't allow writing past the end of the buffer.
    ASSERT(regs.const_buffer.cb_pos + sizeof(u32) <= regs.const_buffer.cb_size);
-    VAddr address =
+    boost::optional<VAddr> address =
-        memory_manager.PhysicalToVirtualAddress(buffer_address + regs.const_buffer.cb_pos);
+        memory_manager.GpuToCpuAddress(buffer_address + regs.const_buffer.cb_pos);
-    Memory::Write32(address, value);
+    Memory::Write32(*address, value);
    // Increment the current buffer position.
    regs.const_buffer.cb_pos = regs.const_buffer.cb_pos + 4;
@@ -213,10 +244,10 @@ Texture::TICEntry Maxwell3D::GetTICEntry(u32 tic_index) const {
    GPUVAddr tic_base_address = regs.tic.TICAddress();
    GPUVAddr tic_address_gpu = tic_base_address + tic_index * sizeof(Texture::TICEntry);
-    VAddr tic_address_cpu = memory_manager.PhysicalToVirtualAddress(tic_address_gpu);
+    boost::optional<VAddr> tic_address_cpu = memory_manager.GpuToCpuAddress(tic_address_gpu);
    Texture::TICEntry tic_entry;
-    Memory::ReadBlock(tic_address_cpu, &tic_entry, sizeof(Texture::TICEntry));
+    Memory::ReadBlock(*tic_address_cpu, &tic_entry, sizeof(Texture::TICEntry));
    ASSERT_MSG(tic_entry.header_version == Texture::TICHeaderVersion::BlockLinear ||
                   tic_entry.header_version == Texture::TICHeaderVersion::Pitch,
@@ -243,10 +274,10 @@ Texture::TSCEntry Maxwell3D::GetTSCEntry(u32 tsc_index) const {
    GPUVAddr tsc_base_address = regs.tsc.TSCAddress();
    GPUVAddr tsc_address_gpu = tsc_base_address + tsc_index * sizeof(Texture::TSCEntry);
-    VAddr tsc_address_cpu = memory_manager.PhysicalToVirtualAddress(tsc_address_gpu);
+    boost::optional<VAddr> tsc_address_cpu = memory_manager.GpuToCpuAddress(tsc_address_gpu);
    Texture::TSCEntry tsc_entry;
-    Memory::ReadBlock(tsc_address_cpu, &tsc_entry, sizeof(Texture::TSCEntry));
+    Memory::ReadBlock(*tsc_address_cpu, &tsc_entry, sizeof(Texture::TSCEntry));
    return tsc_entry;
 }
@@ -268,7 +299,7 @@ std::vector<Texture::FullTextureInfo> Maxwell3D::GetStageTextures(Regs::ShaderSt
         current_texture < tex_info_buffer_end; current_texture += sizeof(Texture::TextureHandle)) {
        Texture::TextureHandle tex_handle{
-            Memory::Read32(memory_manager.PhysicalToVirtualAddress(current_texture))};
+            Memory::Read32(*memory_manager.GpuToCpuAddress(current_texture))};
        Texture::FullTextureInfo tex_info{};
        // TODO(Subv): Use the shader to determine which textures are actually accessed.
diff --git a/src/video_core/engines/maxwell_3d.h b/src/video_core/engines/maxwell_3d.h
index d4fcedace..5cf62fb01 100644
--- a/src/video_core/engines/maxwell_3d.h
+++ b/src/video_core/engines/maxwell_3d.h
@@ -31,7 +31,7 @@ public:
    /// Register structure of the Maxwell3D engine.
    /// TODO(Subv): This structure will need to be made bigger as more registers are discovered.
    struct Regs {
-        static constexpr size_t NUM_REGS = 0xE36;
+        static constexpr size_t NUM_REGS = 0xE00;
        static constexpr size_t NumRenderTargets = 8;
        static constexpr size_t NumViewports = 16;
@@ -46,6 +46,29 @@ public:
        enum class QueryMode : u32 {
            Write = 0,
            Sync = 1,
+            // TODO(Subv): It is currently unknown what the difference between method 2 and method 0
+            // is.
+            Write2 = 2,
+        };
+        enum class QueryUnit : u32 {
+            VFetch = 1,
+            VP = 2,
+            Rast = 4,
+            StrmOut = 5,
+            GP = 6,
+            ZCull = 7,
+            Prop = 10,
+            Crop = 15,
+        };
+        enum class QuerySelect : u32 {
+            Zero = 0,
+        };
+        enum class QuerySyncCondition : u32 {
+            NotEqual = 0,
+            GreaterThan = 1,
        };
        enum class ShaderProgram : u32 {
@@ -299,7 +322,15 @@ public:
        union {
            struct {
-                INSERT_PADDING_WORDS(0x200);
+                INSERT_PADDING_WORDS(0x45);
+                struct {
+                    INSERT_PADDING_WORDS(1);
+                    u32 data;
+                    u32 entry;
+                } macros;
+                INSERT_PADDING_WORDS(0x1B8);
                struct {
                    u32 address_high;
@@ -476,7 +507,10 @@ public:
                        u32 raw;
                        BitField<0, 2, QueryMode> mode;
                        BitField<4, 1, u32> fence;
-                        BitField<12, 4, u32> unit;
+                        BitField<12, 4, QueryUnit> unit;
+                        BitField<16, 1, QuerySyncCondition> sync_cond;
+                        BitField<23, 5, QuerySelect> select;
+                        BitField<28, 1, u32> short_query;
                    } query_get;
                    GPUVAddr QueryAddress() const {
@@ -500,6 +534,11 @@ public:
                        return static_cast<GPUVAddr>((static_cast<GPUVAddr>(start_high) << 32) |
                                                     start_low);
                    }
+                    bool IsEnabled() const {
+                        return enable != 0 && StartAddress() != 0;
+                    }
                } vertex_array[NumVertexArrays];
                Blend blend;
@@ -574,7 +613,7 @@ public:
                    u32 size[MaxShaderStage];
                } tex_info_buffers;
-                INSERT_PADDING_WORDS(0x102);
+                INSERT_PADDING_WORDS(0xCC);
            };
            std::array<u32, NUM_REGS> reg_array;
        };
@@ -606,9 +645,6 @@ public:
    /// Write the value to the register identified by method.
    void WriteReg(u32 method, u32 value, u32 remaining_params);
-    /// Uploads the code for a GPU macro program associated with the specified entry.
-    void SubmitMacroCode(u32 entry, std::vector<u32> code);
    /// Returns a list of enabled textures for the specified shader stage.
    std::vector<Texture::FullTextureInfo> GetStageTextures(Regs::ShaderStage stage) const;
@@ -639,6 +675,9 @@ private:
     */
    void CallMacroMethod(u32 method, std::vector<u32> parameters);
+    /// Handles writes to the macro uploading registers.
+    void ProcessMacroUpload(u32 data);
    /// Handles a write to the QUERY_GET register.
    void ProcessQueryGet();
@@ -656,6 +695,7 @@ private:
    static_assert(offsetof(Maxwell3D::Regs, field_name) == position * 4,                           \
                  "Field " #field_name " has invalid position")
+ASSERT_REG_POSITION(macros, 0x45);
 ASSERT_REG_POSITION(rt, 0x200);
 ASSERT_REG_POSITION(viewport_transform[0], 0x280);
 ASSERT_REG_POSITION(viewport, 0x300);
diff --git a/src/video_core/engines/shader_bytecode.h b/src/video_core/engines/shader_bytecode.h
index 5a006aee5..f4d11fa5d 100644
--- a/src/video_core/engines/shader_bytecode.h
+++ b/src/video_core/engines/shader_bytecode.h
@@ -214,6 +214,20 @@ union Instruction {
        BitField<56, 1, u64> neg_b;
    } fsetp;
+    union {
+        BitField<39, 3, u64> pred39;
+        BitField<42, 1, u64> neg_pred;
+        BitField<43, 1, u64> neg_a;
+        BitField<44, 1, u64> abs_b;
+        BitField<45, 2, PredOperation> op;
+        BitField<48, 4, PredCondition> cond;
+        BitField<53, 1, u64> neg_b;
+        BitField<54, 1, u64> abs_a;
+        BitField<52, 1, u64> bf;
+        BitField<55, 1, u64> ftz;
+        BitField<56, 1, u64> neg_imm;
+    } fset;
    BitField<61, 1, u64> is_b_imm;
    BitField<60, 1, u64> is_b_gpr;
    BitField<59, 1, u64> is_c_gpr;
@@ -261,6 +275,9 @@ public:
        I2F_C,
        I2F_R,
        I2F_IMM,
+        I2I_C,
+        I2I_R,
+        I2I_IMM,
        LOP32I,
        MOV_C,
        MOV_R,
@@ -272,6 +289,9 @@ public:
        FSETP_C, // Set Predicate
        FSETP_R,
        FSETP_IMM,
+        FSET_C,
+        FSET_R,
+        FSET_IMM,
        ISETP_C,
        ISETP_IMM,
        ISETP_R,
@@ -283,8 +303,9 @@ public:
        Ffma,
        Flow,
        Memory,
-        FloatPredicate,
+        FloatSet,
-        IntegerPredicate,
+        FloatSetPredicate,
+        IntegerSetPredicate,
        Unknown,
    };
@@ -409,6 +430,9 @@ private:
            INST("0100110010111---", Id::I2F_C, Type::Arithmetic, "I2F_C"),
            INST("0101110010111---", Id::I2F_R, Type::Arithmetic, "I2F_R"),
            INST("0011100-10111---", Id::I2F_IMM, Type::Arithmetic, "I2F_IMM"),
+            INST("0100110011100---", Id::I2I_C, Type::Arithmetic, "I2I_C"),
+            INST("0101110011100---", Id::I2I_R, Type::Arithmetic, "I2I_R"),
+            INST("01110001-1000---", Id::I2I_IMM, Type::Arithmetic, "I2I_IMM"),
            INST("000001----------", Id::LOP32I, Type::Arithmetic, "LOP32I"),
            INST("0100110010011---", Id::MOV_C, Type::Arithmetic, "MOV_C"),
            INST("0101110010011---", Id::MOV_R, Type::Arithmetic, "MOV_R"),
@@ -417,12 +441,15 @@ private:
            INST("0100110000101---", Id::SHR_C, Type::Arithmetic, "SHR_C"),
            INST("0101110000101---", Id::SHR_R, Type::Arithmetic, "SHR_R"),
            INST("0011100-00101---", Id::SHR_IMM, Type::Arithmetic, "SHR_IMM"),
-            INST("010010111011----", Id::FSETP_C, Type::FloatPredicate, "FSETP_C"),
+            INST("01011000--------", Id::FSET_R, Type::FloatSet, "FSET_R"),
-            INST("010110111011----", Id::FSETP_R, Type::FloatPredicate, "FSETP_R"),
+            INST("0100100---------", Id::FSET_C, Type::FloatSet, "FSET_C"),
-            INST("0011011-1011----", Id::FSETP_IMM, Type::FloatPredicate, "FSETP_IMM"),
+            INST("0011000---------", Id::FSET_IMM, Type::FloatSet, "FSET_IMM"),
-            INST("010010110110----", Id::ISETP_C, Type::IntegerPredicate, "ISETP_C"),
+            INST("010010111011----", Id::FSETP_C, Type::FloatSetPredicate, "FSETP_C"),
-            INST("010110110110----", Id::ISETP_R, Type::IntegerPredicate, "ISETP_R"),
+            INST("010110111011----", Id::FSETP_R, Type::FloatSetPredicate, "FSETP_R"),
-            INST("0011011-0110----", Id::ISETP_IMM, Type::IntegerPredicate, "ISETP_IMM"),
+            INST("0011011-1011----", Id::FSETP_IMM, Type::FloatSetPredicate, "FSETP_IMM"),
+            INST("010010110110----", Id::ISETP_C, Type::IntegerSetPredicate, "ISETP_C"),
+            INST("010110110110----", Id::ISETP_R, Type::IntegerSetPredicate, "ISETP_R"),
+            INST("0011011-0110----", Id::ISETP_IMM, Type::IntegerSetPredicate, "ISETP_IMM"),
        };
 #undef INST
        std::stable_sort(table.begin(), table.end(), [](const auto& a, const auto& b) {
diff --git a/src/video_core/gpu.cpp b/src/video_core/gpu.cpp
index 9463cd5d6..9eb143918 100644
--- a/src/video_core/gpu.cpp
+++ b/src/video_core/gpu.cpp
@@ -12,7 +12,7 @@ namespace Tegra {
 GPU::GPU() {
    memory_manager = std::make_unique<MemoryManager>();
    maxwell_3d = std::make_unique<Engines::Maxwell3D>(*memory_manager);
-    fermi_2d = std::make_unique<Engines::Fermi2D>();
+    fermi_2d = std::make_unique<Engines::Fermi2D>(*memory_manager);
    maxwell_compute = std::make_unique<Engines::MaxwellCompute>();
 }
@@ -22,4 +22,16 @@ const Tegra::Engines::Maxwell3D& GPU::Get3DEngine() const {
    return *maxwell_3d;
 }
+u32 RenderTargetBytesPerPixel(RenderTargetFormat format) {
+    ASSERT(format != RenderTargetFormat::NONE);
+    switch (format) {
+    case RenderTargetFormat::RGBA8_UNORM:
+    case RenderTargetFormat::RGB10_A2_UNORM:
+        return 4;
+    default:
+        UNIMPLEMENTED_MSG("Unimplemented render target format %u", static_cast<u32>(format));
+    }
+}
 } // namespace Tegra
diff --git a/src/video_core/gpu.h b/src/video_core/gpu.h
index 2888daedc..f168a5171 100644
--- a/src/video_core/gpu.h
+++ b/src/video_core/gpu.h
@@ -21,6 +21,9 @@ enum class RenderTargetFormat : u32 {
    RGBA8_SRGB = 0xD6,
 };
+/// Returns the number of bytes per pixel of each rendertarget format.
+u32 RenderTargetBytesPerPixel(RenderTargetFormat format);
 class DebugContext;
 /**
@@ -86,8 +89,6 @@ public:
    }
 private:
-    static constexpr u32 InvalidGraphMacroEntry = 0xFFFFFFFF;
    /// Writes a single register in the engine bound to the specified subchannel
    void WriteReg(u32 method, u32 subchannel, u32 value, u32 remaining_params);
@@ -100,11 +101,6 @@ private:
    std::unique_ptr<Engines::Fermi2D> fermi_2d;
    /// Compute engine
    std::unique_ptr<Engines::MaxwellCompute> maxwell_compute;
-    /// Entry of the macro that is currently being uploaded
-    u32 current_macro_entry = InvalidGraphMacroEntry;
-    /// Code being uploaded for the current macro
-    std::vector<u32> current_macro_code;
 };
 } // namespace Tegra
diff --git a/src/video_core/memory_manager.cpp b/src/video_core/memory_manager.cpp
index 2789a4ca1..25984439d 100644
--- a/src/video_core/memory_manager.cpp
+++ b/src/video_core/memory_manager.cpp
@@ -2,109 +2,118 @@
 // Licensed under GPLv2 or any later version
 // Refer to the license.txt file included.
+#include "common/alignment.h"
 #include "common/assert.h"
 #include "video_core/memory_manager.h"
 namespace Tegra {
-PAddr MemoryManager::AllocateSpace(u64 size, u64 align) {
+GPUVAddr MemoryManager::AllocateSpace(u64 size, u64 align) {
-    boost::optional<PAddr> paddr = FindFreeBlock(size, align);
+    boost::optional<GPUVAddr> gpu_addr = FindFreeBlock(size, align);
-    ASSERT(paddr);
+    ASSERT(gpu_addr);
-    for (u64 offset = 0; offset < size; offset += Memory::PAGE_SIZE) {
+    for (u64 offset = 0; offset < size; offset += PAGE_SIZE) {
-        PageSlot(*paddr + offset) = static_cast<u64>(PageStatus::Allocated);
+        ASSERT(PageSlot(*gpu_addr + offset) == static_cast<u64>(PageStatus::Unmapped));
+        PageSlot(*gpu_addr + offset) = static_cast<u64>(PageStatus::Allocated);
    }
-    return *paddr;
+    return *gpu_addr;
 }
-PAddr MemoryManager::AllocateSpace(PAddr paddr, u64 size, u64 align) {
+GPUVAddr MemoryManager::AllocateSpace(GPUVAddr gpu_addr, u64 size, u64 align) {
-    for (u64 offset = 0; offset < size; offset += Memory::PAGE_SIZE) {
+    for (u64 offset = 0; offset < size; offset += PAGE_SIZE) {
-        if (IsPageMapped(paddr + offset)) {
+        ASSERT(PageSlot(gpu_addr + offset) == static_cast<u64>(PageStatus::Unmapped));
-            return AllocateSpace(size, align);
+        PageSlot(gpu_addr + offset) = static_cast<u64>(PageStatus::Allocated);
-        }
-    }
-    for (u64 offset = 0; offset < size; offset += Memory::PAGE_SIZE) {
-        PageSlot(paddr + offset) = static_cast<u64>(PageStatus::Allocated);
    }
-    return paddr;
+    return gpu_addr;
 }
-PAddr MemoryManager::MapBufferEx(VAddr vaddr, u64 size) {
+GPUVAddr MemoryManager::MapBufferEx(VAddr cpu_addr, u64 size) {
-    vaddr &= ~Memory::PAGE_MASK;
+    boost::optional<GPUVAddr> gpu_addr = FindFreeBlock(size, PAGE_SIZE);
+    ASSERT(gpu_addr);
-    boost::optional<PAddr> paddr = FindFreeBlock(size);
-    ASSERT(paddr);
-    for (u64 offset = 0; offset < size; offset += Memory::PAGE_SIZE) {
+    for (u64 offset = 0; offset < size; offset += PAGE_SIZE) {
-        PageSlot(*paddr + offset) = vaddr + offset;
+        ASSERT(PageSlot(*gpu_addr + offset) == static_cast<u64>(PageStatus::Unmapped));
+        PageSlot(*gpu_addr + offset) = cpu_addr + offset;
    }
-    return *paddr;
+    MappedRegion region{cpu_addr, *gpu_addr, size};
+    mapped_regions.push_back(region);
+    return *gpu_addr;
 }
-PAddr MemoryManager::MapBufferEx(VAddr vaddr, PAddr paddr, u64 size) {
+GPUVAddr MemoryManager::MapBufferEx(VAddr cpu_addr, GPUVAddr gpu_addr, u64 size) {
-    vaddr &= ~Memory::PAGE_MASK;
+    ASSERT((gpu_addr & PAGE_MASK) == 0);
-    paddr &= ~Memory::PAGE_MASK;
-    for (u64 offset = 0; offset < size; offset += Memory::PAGE_SIZE) {
+    for (u64 offset = 0; offset < size; offset += PAGE_SIZE) {
-        if (PageSlot(paddr + offset) != static_cast<u64>(PageStatus::Allocated)) {
+        ASSERT(PageSlot(gpu_addr + offset) == static_cast<u64>(PageStatus::Allocated));
-            return MapBufferEx(vaddr, size);
+        PageSlot(gpu_addr + offset) = cpu_addr + offset;
-        }
    }
-    for (u64 offset = 0; offset < size; offset += Memory::PAGE_SIZE) {
+    MappedRegion region{cpu_addr, gpu_addr, size};
-        PageSlot(paddr + offset) = vaddr + offset;
+    mapped_regions.push_back(region);
-    }
-    return paddr;
+    return gpu_addr;
 }
-boost::optional<PAddr> MemoryManager::FindFreeBlock(u64 size, u64 align) {
+boost::optional<GPUVAddr> MemoryManager::FindFreeBlock(u64 size, u64 align) {
-    PAddr paddr{};
+    GPUVAddr gpu_addr = 0;
-    u64 free_space{};
+    u64 free_space = 0;
-    align = (align + Memory::PAGE_MASK) & ~Memory::PAGE_MASK;
+    align = (align + PAGE_MASK) & ~PAGE_MASK;
-    while (paddr + free_space < MAX_ADDRESS) {
+    while (gpu_addr + free_space < MAX_ADDRESS) {
-        if (!IsPageMapped(paddr + free_space)) {
+        if (!IsPageMapped(gpu_addr + free_space)) {
-            free_space += Memory::PAGE_SIZE;
+            free_space += PAGE_SIZE;
            if (free_space >= size) {
-                return paddr;
+                return gpu_addr;
            }
        } else {
-            paddr += free_space + Memory::PAGE_SIZE;
+            gpu_addr += free_space + PAGE_SIZE;
            free_space = 0;
-            const u64 remainder{paddr % align};
+            gpu_addr = Common::AlignUp(gpu_addr, align);
-            if (!remainder) {
-                paddr = (paddr - remainder) + align;
-            }
        }
    }
    return {};
 }
-VAddr MemoryManager::PhysicalToVirtualAddress(PAddr paddr) {
+boost::optional<VAddr> MemoryManager::GpuToCpuAddress(GPUVAddr gpu_addr) {
-    VAddr base_addr = PageSlot(paddr);
+    VAddr base_addr = PageSlot(gpu_addr);
    ASSERT(base_addr != static_cast<u64>(PageStatus::Unmapped));
-    return base_addr + (paddr & Memory::PAGE_MASK);
+    if (base_addr == static_cast<u64>(PageStatus::Allocated)) {
+        return {};
+    }
+    return base_addr + (gpu_addr & PAGE_MASK);
+}
+std::vector<GPUVAddr> MemoryManager::CpuToGpuAddress(VAddr cpu_addr) const {
+    std::vector<GPUVAddr> results;
+    for (const auto& region : mapped_regions) {
+        if (cpu_addr >= region.cpu_addr && cpu_addr < (region.cpu_addr + region.size)) {
+            u64 offset = cpu_addr - region.cpu_addr;
+            results.push_back(region.gpu_addr + offset);
+        }
+    }
+    return results;
 }
-bool MemoryManager::IsPageMapped(PAddr paddr) {
+bool MemoryManager::IsPageMapped(GPUVAddr gpu_addr) {
-    return PageSlot(paddr) != static_cast<u64>(PageStatus::Unmapped);
+    return PageSlot(gpu_addr) != static_cast<u64>(PageStatus::Unmapped);
 }
-VAddr& MemoryManager::PageSlot(PAddr paddr) {
+VAddr& MemoryManager::PageSlot(GPUVAddr gpu_addr) {
-    auto& block = page_table[(paddr >> (Memory::PAGE_BITS + PAGE_TABLE_BITS)) & PAGE_TABLE_MASK];
+    auto& block = page_table[(gpu_addr >> (PAGE_BITS + PAGE_TABLE_BITS)) & PAGE_TABLE_MASK];
    if (!block) {
        block = std::make_unique<PageBlock>();
        for (unsigned index = 0; index < PAGE_BLOCK_SIZE; index++) {
            (*block)[index] = static_cast<u64>(PageStatus::Unmapped);
        }
    }
-    return (*block)[(paddr >> Memory::PAGE_BITS) & PAGE_BLOCK_MASK];
+    return (*block)[(gpu_addr >> PAGE_BITS) & PAGE_BLOCK_MASK];
 }
 } // namespace Tegra
diff --git a/src/video_core/memory_manager.h b/src/video_core/memory_manager.h
index 47da7acd6..08140c83a 100644
--- a/src/video_core/memory_manager.h
+++ b/src/video_core/memory_manager.h
@@ -6,8 +6,11 @@
 #include <array>
 #include <memory>
+#include <vector>
+#include <boost/optional.hpp>
 #include "common/common_types.h"
-#include "core/memory.h"
 namespace Tegra {
@@ -18,16 +21,21 @@ class MemoryManager final {
 public:
    MemoryManager() = default;
-    PAddr AllocateSpace(u64 size, u64 align);
+    GPUVAddr AllocateSpace(u64 size, u64 align);
-    PAddr AllocateSpace(PAddr paddr, u64 size, u64 align);
+    GPUVAddr AllocateSpace(GPUVAddr gpu_addr, u64 size, u64 align);
-    PAddr MapBufferEx(VAddr vaddr, u64 size);
+    GPUVAddr MapBufferEx(VAddr cpu_addr, u64 size);
-    PAddr MapBufferEx(VAddr vaddr, PAddr paddr, u64 size);
+    GPUVAddr MapBufferEx(VAddr cpu_addr, GPUVAddr gpu_addr, u64 size);
-    VAddr PhysicalToVirtualAddress(PAddr paddr);
+    boost::optional<VAddr> GpuToCpuAddress(GPUVAddr gpu_addr);
+    std::vector<GPUVAddr> CpuToGpuAddress(VAddr cpu_addr) const;
+    static constexpr u64 PAGE_BITS = 16;
+    static constexpr u64 PAGE_SIZE = 1 << PAGE_BITS;
+    static constexpr u64 PAGE_MASK = PAGE_SIZE - 1;
 private:
-    boost::optional<PAddr> FindFreeBlock(u64 size, u64 align = 1);
+    boost::optional<GPUVAddr> FindFreeBlock(u64 size, u64 align = 1);
-    bool IsPageMapped(PAddr paddr);
+    bool IsPageMapped(GPUVAddr gpu_addr);
-    VAddr& PageSlot(PAddr paddr);
+    VAddr& PageSlot(GPUVAddr gpu_addr);
    enum class PageStatus : u64 {
        Unmapped = 0xFFFFFFFFFFFFFFFFULL,
@@ -35,7 +43,7 @@ private:
    };
    static constexpr u64 MAX_ADDRESS{0x10000000000ULL};
-    static constexpr u64 PAGE_TABLE_BITS{14};
+    static constexpr u64 PAGE_TABLE_BITS{10};
    static constexpr u64 PAGE_TABLE_SIZE{1 << PAGE_TABLE_BITS};
    static constexpr u64 PAGE_TABLE_MASK{PAGE_TABLE_SIZE - 1};
    static constexpr u64 PAGE_BLOCK_BITS{14};
@@ -44,6 +52,14 @@ private:
    using PageBlock = std::array<VAddr, PAGE_BLOCK_SIZE>;
    std::array<std::unique_ptr<PageBlock>, PAGE_TABLE_SIZE> page_table{};
+    struct MappedRegion {
+        VAddr cpu_addr;
+        GPUVAddr gpu_addr;
+        u64 size;
+    };
+    std::vector<MappedRegion> mapped_regions;
 };
 } // namespace Tegra
diff --git a/src/video_core/rasterizer_interface.h b/src/video_core/rasterizer_interface.h
index 36629dd11..f0e48a802 100644
--- a/src/video_core/rasterizer_interface.h
+++ b/src/video_core/rasterizer_interface.h
@@ -6,6 +6,7 @@
 #include "common/common_types.h"
 #include "video_core/gpu.h"
+#include "video_core/memory_manager.h"
 struct ScreenInfo;
@@ -25,14 +26,14 @@ public:
    virtual void FlushAll() = 0;
    /// Notify rasterizer that any caches of the specified region should be flushed to Switch memory
-    virtual void FlushRegion(VAddr addr, u64 size) = 0;
+    virtual void FlushRegion(Tegra::GPUVAddr addr, u64 size) = 0;
    /// Notify rasterizer that any caches of the specified region should be invalidated
-    virtual void InvalidateRegion(VAddr addr, u64 size) = 0;
+    virtual void InvalidateRegion(Tegra::GPUVAddr addr, u64 size) = 0;
    /// Notify rasterizer that any caches of the specified region should be flushed to Switch memory
    /// and invalidated
-    virtual void FlushAndInvalidateRegion(VAddr addr, u64 size) = 0;
+    virtual void FlushAndInvalidateRegion(Tegra::GPUVAddr addr, u64 size) = 0;
    /// Attempt to use a faster method to perform a display transfer with is_texture_copy = 0
    virtual bool AccelerateDisplayTransfer(const void* config) {
diff --git a/src/video_core/renderer_opengl/gl_rasterizer.cpp b/src/video_core/renderer_opengl/gl_rasterizer.cpp
index 2d4a0d6db..9b3542e10 100644
--- a/src/video_core/renderer_opengl/gl_rasterizer.cpp
+++ b/src/video_core/renderer_opengl/gl_rasterizer.cpp
@@ -116,7 +116,7 @@ RasterizerOpenGL::RasterizerOpenGL() {
    glEnable(GL_BLEND);
-    LOG_CRITICAL(Render_OpenGL, "Sync fixed function OpenGL state here!");
+    NGLOG_CRITICAL(Render_OpenGL, "Sync fixed function OpenGL state here!");
 }
 RasterizerOpenGL::~RasterizerOpenGL() {
@@ -127,7 +127,8 @@ RasterizerOpenGL::~RasterizerOpenGL() {
    }
 }
-void RasterizerOpenGL::SetupVertexArray(u8* array_ptr, GLintptr buffer_offset) {
+std::pair<u8*, GLintptr> RasterizerOpenGL::SetupVertexArrays(u8* array_ptr,
+                                                             GLintptr buffer_offset) {
    MICROPROFILE_SCOPE(OpenGL_VAO);
    const auto& regs = Core::System().GetInstance().GPU().Maxwell3D().regs;
    const auto& memory_manager = Core::System().GetInstance().GPU().memory_manager;
@@ -136,43 +137,58 @@ void RasterizerOpenGL::SetupVertexArray(u8* array_ptr, GLintptr buffer_offset) {
    state.draw.vertex_buffer = stream_buffer->GetHandle();
    state.Apply();
-    // TODO(bunnei): Add support for 1+ vertex arrays
+    // Upload all guest vertex arrays sequentially to our buffer
-    const auto& vertex_array{regs.vertex_array[0]};
+    for (u32 index = 0; index < Maxwell::NumVertexArrays; ++index) {
-    const auto& vertex_array_limit{regs.vertex_array_limit[0]};
+        const auto& vertex_array = regs.vertex_array[index];
-    ASSERT_MSG(vertex_array.enable, "vertex array 0 is disabled?");
+        if (!vertex_array.IsEnabled())
-    ASSERT_MSG(!vertex_array.divisor, "vertex array 0 divisor is unimplemented!");
+            continue;
-    for (unsigned index = 1; index < Maxwell::NumVertexArrays; ++index) {
-        ASSERT_MSG(!regs.vertex_array[index].enable, "vertex array %d is unimplemented!", index);
+        const Tegra::GPUVAddr start = vertex_array.StartAddress();
+        const Tegra::GPUVAddr end = regs.vertex_array_limit[index].LimitAddress();
+        ASSERT(end > start);
+        u64 size = end - start + 1;
+        // Copy vertex array data
+        res_cache.FlushRegion(start, size, nullptr);
+        Memory::ReadBlock(*memory_manager->GpuToCpuAddress(start), array_ptr, size);
+        // Bind the vertex array to the buffer at the current offset.
+        glBindVertexBuffer(index, stream_buffer->GetHandle(), buffer_offset, vertex_array.stride);
+        ASSERT_MSG(vertex_array.divisor == 0, "Vertex buffer divisor unimplemented");
+        array_ptr += size;
+        buffer_offset += size;
    }
    // Use the vertex array as-is, assumes that the data is formatted correctly for OpenGL.
    // Enables the first 16 vertex attributes always, as we don't know which ones are actually used
-    // until shader time. Note, Tegra technically supports 32, but we're cappinig this to 16 for now
+    // until shader time. Note, Tegra technically supports 32, but we're capping this to 16 for now
    // to avoid OpenGL errors.
+    // TODO(Subv): Analyze the shader to identify which attributes are actually used and don't
+    // assume every shader uses them all.
    for (unsigned index = 0; index < 16; ++index) {
        auto& attrib = regs.vertex_attrib_format[index];
        NGLOG_DEBUG(HW_GPU, "vertex attrib {}, count={}, size={}, type={}, offset={}, normalize={}",
                    index, attrib.ComponentCount(), attrib.SizeString(), attrib.TypeString(),
                    attrib.offset.Value(), attrib.IsNormalized());
-        glVertexAttribPointer(index, attrib.ComponentCount(), MaxwellToGL::VertexType(attrib),
+        auto& buffer = regs.vertex_array[attrib.buffer];
-                              attrib.IsNormalized() ? GL_TRUE : GL_FALSE, vertex_array.stride,
+        ASSERT(buffer.IsEnabled());
-                              reinterpret_cast<GLvoid*>(buffer_offset + attrib.offset));
        glEnableVertexAttribArray(index);
+        glVertexAttribFormat(index, attrib.ComponentCount(), MaxwellToGL::VertexType(attrib),
+                             attrib.IsNormalized() ? GL_TRUE : GL_FALSE, attrib.offset);
+        glVertexAttribBinding(index, attrib.buffer);
        hw_vao_enabled_attributes[index] = true;
    }
-    // Copy vertex array data
+    return {array_ptr, buffer_offset};
-    const u64 data_size{vertex_array_limit.LimitAddress() - vertex_array.StartAddress() + 1};
-    const VAddr data_addr{memory_manager->PhysicalToVirtualAddress(vertex_array.StartAddress())};
-    res_cache.FlushRegion(data_addr, data_size, nullptr);
-    Memory::ReadBlock(data_addr, array_ptr, data_size);
-    array_ptr += data_size;
-    buffer_offset += data_size;
 }
-void RasterizerOpenGL::SetupShaders(u8* buffer_ptr, GLintptr buffer_offset, size_t ptr_pos) {
+void RasterizerOpenGL::SetupShaders(u8* buffer_ptr, GLintptr buffer_offset) {
    // Helper function for uploading uniform data
    const auto copy_buffer = [&](GLuint handle, GLintptr offset, GLsizeiptr size) {
        if (has_ARB_direct_state_access) {
@@ -190,8 +206,6 @@ void RasterizerOpenGL::SetupShaders(u8* buffer_ptr, GLintptr buffer_offset, size
    u32 current_constbuffer_bindpoint = 0;
    for (unsigned index = 1; index < Maxwell::MaxShaderProgram; ++index) {
-        ptr_pos += sizeof(GLShader::MaxwellUniformData);
        auto& shader_config = gpu.regs.shader_config[index];
        const Maxwell::ShaderProgram program{static_cast<Maxwell::ShaderProgram>(index)};
@@ -205,18 +219,21 @@ void RasterizerOpenGL::SetupShaders(u8* buffer_ptr, GLintptr buffer_offset, size
        }
        // Upload uniform data as one UBO per stage
-        const GLintptr ubo_offset = buffer_offset + static_cast<GLintptr>(ptr_pos);
+        const GLintptr ubo_offset = buffer_offset;
        copy_buffer(uniform_buffers[stage].handle, ubo_offset,
                    sizeof(GLShader::MaxwellUniformData));
        GLShader::MaxwellUniformData* ub_ptr =
-            reinterpret_cast<GLShader::MaxwellUniformData*>(&buffer_ptr[ptr_pos]);
+            reinterpret_cast<GLShader::MaxwellUniformData*>(buffer_ptr);
        ub_ptr->SetFromRegs(gpu.state.shader_stages[stage]);
+        buffer_ptr += sizeof(GLShader::MaxwellUniformData);
+        buffer_offset += sizeof(GLShader::MaxwellUniformData);
        // Fetch program code from memory
        GLShader::ProgramCode program_code;
        const u64 gpu_address{gpu.regs.code_address.CodeAddress() + shader_config.offset};
-        const VAddr cpu_address{gpu.memory_manager.PhysicalToVirtualAddress(gpu_address)};
+        const boost::optional<VAddr> cpu_address{gpu.memory_manager.GpuToCpuAddress(gpu_address)};
-        Memory::ReadBlock(cpu_address, program_code.data(), program_code.size() * sizeof(u64));
+        Memory::ReadBlock(*cpu_address, program_code.data(), program_code.size() * sizeof(u64));
        GLShader::ShaderSetup setup{std::move(program_code)};
        GLShader::ShaderEntries shader_resources;
@@ -235,8 +252,8 @@ void RasterizerOpenGL::SetupShaders(u8* buffer_ptr, GLintptr buffer_offset, size
            break;
        }
        default:
-            LOG_CRITICAL(HW_GPU, "Unimplemented shader index=%d, enable=%d, offset=0x%08X", index,
+            NGLOG_CRITICAL(HW_GPU, "Unimplemented shader index={}, enable={}, offset={:#010X}",
-                         shader_config.enable.Value(), shader_config.offset);
+                           index, shader_config.enable.Value(), shader_config.offset);
            UNREACHABLE();
        }
@@ -252,6 +269,24 @@ void RasterizerOpenGL::SetupShaders(u8* buffer_ptr, GLintptr buffer_offset, size
    shader_program_manager->UseTrivialGeometryShader();
 }
+size_t RasterizerOpenGL::CalculateVertexArraysSize() const {
+    const auto& regs = Core::System().GetInstance().GPU().Maxwell3D().regs;
+    size_t size = 0;
+    for (u32 index = 0; index < Maxwell::NumVertexArrays; ++index) {
+        if (!regs.vertex_array[index].IsEnabled())
+            continue;
+        const Tegra::GPUVAddr start = regs.vertex_array[index].StartAddress();
+        const Tegra::GPUVAddr end = regs.vertex_array_limit[index].LimitAddress();
+        ASSERT(end > start);
+        size += end - start + 1;
+    }
+    return size;
+}
 bool RasterizerOpenGL::AccelerateDrawBatch(bool is_indexed) {
    accelerate_draw = is_indexed ? AccelDraw::Indexed : AccelDraw::Arrays;
    DrawArrays();
@@ -329,44 +364,49 @@ void RasterizerOpenGL::DrawArrays() {
    const u64 index_buffer_size{regs.index_array.count * regs.index_array.FormatSizeInBytes()};
    const unsigned vertex_num{is_indexed ? regs.index_array.count : regs.vertex_buffer.count};
-    // TODO(bunnei): Add support for 1+ vertex arrays
-    vs_input_size = vertex_num * regs.vertex_array[0].stride;
    state.draw.vertex_buffer = stream_buffer->GetHandle();
    state.Apply();
-    size_t buffer_size = static_cast<size_t>(vs_input_size);
+    size_t buffer_size = CalculateVertexArraysSize();
    if (is_indexed) {
-        buffer_size = Common::AlignUp(buffer_size, 4) + index_buffer_size;
+        buffer_size = Common::AlignUp<size_t>(buffer_size, 4) + index_buffer_size;
    }
    // Uniform space for the 5 shader stages
-    buffer_size += sizeof(GLShader::MaxwellUniformData) * Maxwell::MaxShaderStage;
+    buffer_size = Common::AlignUp<size_t>(buffer_size, 4) +
+                  sizeof(GLShader::MaxwellUniformData) * Maxwell::MaxShaderStage;
-    size_t ptr_pos = 0;
    u8* buffer_ptr;
    GLintptr buffer_offset;
    std::tie(buffer_ptr, buffer_offset) =
        stream_buffer->Map(static_cast<GLsizeiptr>(buffer_size), 4);
-    SetupVertexArray(buffer_ptr, buffer_offset);
+    u8* offseted_buffer;
-    ptr_pos += vs_input_size;
+    std::tie(offseted_buffer, buffer_offset) = SetupVertexArrays(buffer_ptr, buffer_offset);
+    offseted_buffer =
+        reinterpret_cast<u8*>(Common::AlignUp(reinterpret_cast<size_t>(offseted_buffer), 4));
+    buffer_offset = Common::AlignUp<size_t>(buffer_offset, 4);
    // If indexed mode, copy the index buffer
    GLintptr index_buffer_offset = 0;
    if (is_indexed) {
-        ptr_pos = Common::AlignUp(ptr_pos, 4);
        const auto& memory_manager = Core::System().GetInstance().GPU().memory_manager;
-        const VAddr index_data_addr{
+        const boost::optional<VAddr> index_data_addr{
-            memory_manager->PhysicalToVirtualAddress(regs.index_array.StartAddress())};
+            memory_manager->GpuToCpuAddress(regs.index_array.StartAddress())};
-        Memory::ReadBlock(index_data_addr, &buffer_ptr[ptr_pos], index_buffer_size);
+        Memory::ReadBlock(*index_data_addr, offseted_buffer, index_buffer_size);
-        index_buffer_offset = buffer_offset + static_cast<GLintptr>(ptr_pos);
+        index_buffer_offset = buffer_offset;
-        ptr_pos += index_buffer_size;
+        offseted_buffer += index_buffer_size;
+        buffer_offset += index_buffer_size;
    }
-    SetupShaders(buffer_ptr, buffer_offset, ptr_pos);
+    offseted_buffer =
+        reinterpret_cast<u8*>(Common::AlignUp(reinterpret_cast<size_t>(offseted_buffer), 4));
+    buffer_offset = Common::AlignUp<size_t>(buffer_offset, 4);
+    SetupShaders(offseted_buffer, buffer_offset);
    stream_buffer->Unmap();
@@ -478,17 +518,17 @@ void RasterizerOpenGL::FlushAll() {
    res_cache.FlushAll();
 }
-void RasterizerOpenGL::FlushRegion(VAddr addr, u64 size) {
+void RasterizerOpenGL::FlushRegion(Tegra::GPUVAddr addr, u64 size) {
    MICROPROFILE_SCOPE(OpenGL_CacheManagement);
    res_cache.FlushRegion(addr, size);
 }
-void RasterizerOpenGL::InvalidateRegion(VAddr addr, u64 size) {
+void RasterizerOpenGL::InvalidateRegion(Tegra::GPUVAddr addr, u64 size) {
    MICROPROFILE_SCOPE(OpenGL_CacheManagement);
    res_cache.InvalidateRegion(addr, size, nullptr);
 }
-void RasterizerOpenGL::FlushAndInvalidateRegion(VAddr addr, u64 size) {
+void RasterizerOpenGL::FlushAndInvalidateRegion(Tegra::GPUVAddr addr, u64 size) {
    MICROPROFILE_SCOPE(OpenGL_CacheManagement);
    res_cache.FlushRegion(addr, size);
    res_cache.InvalidateRegion(addr, size, nullptr);
@@ -519,7 +559,8 @@ bool RasterizerOpenGL::AccelerateDisplay(const Tegra::FramebufferConfig& framebu
    MICROPROFILE_SCOPE(OpenGL_CacheManagement);
    SurfaceParams src_params;
-    src_params.addr = framebuffer_addr;
+    src_params.cpu_addr = framebuffer_addr;
+    src_params.addr = res_cache.TryFindFramebufferGpuAddress(framebuffer_addr).get_value_or(0);
    src_params.width = std::min(framebuffer.width, pixel_stride);
    src_params.height = framebuffer.height;
    src_params.stride = pixel_stride;
@@ -618,9 +659,9 @@ u32 RasterizerOpenGL::SetupConstBuffers(Maxwell::ShaderStage stage, GLuint progr
        buffer_draw_state.enabled = true;
        buffer_draw_state.bindpoint = current_bindpoint + bindpoint;
-        VAddr addr = gpu.memory_manager->PhysicalToVirtualAddress(buffer.address);
+        boost::optional<VAddr> addr = gpu.memory_manager->GpuToCpuAddress(buffer.address);
        std::vector<u8> data(used_buffer.GetSize() * sizeof(float));
-        Memory::ReadBlock(addr, data.data(), data.size());
+        Memory::ReadBlock(*addr, data.data(), data.size());
        glBindBuffer(GL_SHADER_STORAGE_BUFFER, buffer_draw_state.ssbo);
        glBufferData(GL_SHADER_STORAGE_BUFFER, data.size(), data.data(), GL_DYNAMIC_DRAW);
diff --git a/src/video_core/renderer_opengl/gl_rasterizer.h b/src/video_core/renderer_opengl/gl_rasterizer.h
index 03e02b52a..9709e595e 100644
--- a/src/video_core/renderer_opengl/gl_rasterizer.h
+++ b/src/video_core/renderer_opengl/gl_rasterizer.h
@@ -11,6 +11,7 @@
 #include <glad/glad.h>
 #include "common/common_types.h"
 #include "video_core/engines/maxwell_3d.h"
+#include "video_core/memory_manager.h"
 #include "video_core/rasterizer_interface.h"
 #include "video_core/renderer_opengl/gl_rasterizer_cache.h"
 #include "video_core/renderer_opengl/gl_resource_manager.h"
@@ -29,9 +30,9 @@ public:
    void DrawArrays() override;
    void NotifyMaxwellRegisterChanged(u32 method) override;
    void FlushAll() override;
-    void FlushRegion(VAddr addr, u64 size) override;
+    void FlushRegion(Tegra::GPUVAddr addr, u64 size) override;
-    void InvalidateRegion(VAddr addr, u64 size) override;
+    void InvalidateRegion(Tegra::GPUVAddr addr, u64 size) override;
-    void FlushAndInvalidateRegion(VAddr addr, u64 size) override;
+    void FlushAndInvalidateRegion(Tegra::GPUVAddr addr, u64 size) override;
    bool AccelerateDisplayTransfer(const void* config) override;
    bool AccelerateTextureCopy(const void* config) override;
    bool AccelerateFill(const void* config) override;
@@ -148,13 +149,13 @@ private:
    static constexpr size_t STREAM_BUFFER_SIZE = 4 * 1024 * 1024;
    std::unique_ptr<OGLStreamBuffer> stream_buffer;
-    GLsizeiptr vs_input_size;
+    size_t CalculateVertexArraysSize() const;
-    void SetupVertexArray(u8* array_ptr, GLintptr buffer_offset);
+    std::pair<u8*, GLintptr> SetupVertexArrays(u8* array_ptr, GLintptr buffer_offset);
    std::array<OGLBuffer, Tegra::Engines::Maxwell3D::Regs::MaxShaderStage> uniform_buffers;
-    void SetupShaders(u8* buffer_ptr, GLintptr buffer_offset, size_t ptr_pos);
+    void SetupShaders(u8* buffer_ptr, GLintptr buffer_offset);
    enum class AccelDraw { Disabled, Arrays, Indexed };
    AccelDraw accelerate_draw;
diff --git a/src/video_core/renderer_opengl/gl_rasterizer_cache.cpp b/src/video_core/renderer_opengl/gl_rasterizer_cache.cpp
index 7410471cc..501d15e98 100644
--- a/src/video_core/renderer_opengl/gl_rasterizer_cache.cpp
+++ b/src/video_core/renderer_opengl/gl_rasterizer_cache.cpp
@@ -41,18 +41,15 @@ struct FormatTuple {
    GLenum format;
    GLenum type;
    bool compressed;
-    // How many pixels in the original texture are equivalent to one pixel in the compressed
-    // texture.
-    u32 compression_factor;
 };
 static constexpr std::array<FormatTuple, SurfaceParams::MaxPixelFormat> tex_format_tuples = {{
-    {GL_RGBA8, GL_RGBA, GL_UNSIGNED_INT_8_8_8_8_REV, false, 1},                     // ABGR8
+    {GL_RGBA8, GL_RGBA, GL_UNSIGNED_INT_8_8_8_8_REV, false},                    // ABGR8
-    {GL_RGB, GL_RGB, GL_UNSIGNED_SHORT_5_6_5_REV, false, 1},                        // B5G6R5
+    {GL_RGB, GL_RGB, GL_UNSIGNED_SHORT_5_6_5_REV, false},                       // B5G6R5
-    {GL_RGB10_A2, GL_RGBA, GL_UNSIGNED_INT_2_10_10_10_REV, false, 1},               // A2B10G10R10
+    {GL_RGB10_A2, GL_RGBA, GL_UNSIGNED_INT_2_10_10_10_REV, false},              // A2B10G10R10
-    {GL_COMPRESSED_RGB_S3TC_DXT1_EXT, GL_RGB, GL_UNSIGNED_INT_8_8_8_8, true, 16},   // DXT1
+    {GL_COMPRESSED_RGB_S3TC_DXT1_EXT, GL_RGB, GL_UNSIGNED_INT_8_8_8_8, true},   // DXT1
-    {GL_COMPRESSED_RGBA_S3TC_DXT3_EXT, GL_RGBA, GL_UNSIGNED_INT_8_8_8_8, true, 16}, // DXT23
+    {GL_COMPRESSED_RGBA_S3TC_DXT3_EXT, GL_RGBA, GL_UNSIGNED_INT_8_8_8_8, true}, // DXT23
-    {GL_COMPRESSED_RGBA_S3TC_DXT5_EXT, GL_RGBA, GL_UNSIGNED_INT_8_8_8_8, true, 16}, // DXT45
+    {GL_COMPRESSED_RGBA_S3TC_DXT5_EXT, GL_RGBA, GL_UNSIGNED_INT_8_8_8_8, true}, // DXT45
 }};
 static const FormatTuple& GetFormatTuple(PixelFormat pixel_format, ComponentType component_type) {
@@ -83,26 +80,30 @@ static u16 GetResolutionScaleFactor() {
 }
 template <bool morton_to_gl, PixelFormat format>
-void MortonCopy(u32 stride, u32 block_height, u32 height, u8* gl_buffer, VAddr base, VAddr start,
+void MortonCopy(u32 stride, u32 block_height, u32 height, u8* gl_buffer, Tegra::GPUVAddr base,
-                VAddr end) {
+                Tegra::GPUVAddr start, Tegra::GPUVAddr end) {
-    constexpr u32 bytes_per_pixel = SurfaceParams::GetFormatBpp(format) / 8;
+    constexpr u32 bytes_per_pixel = SurfaceParams::GetFormatBpp(format) / CHAR_BIT;
    constexpr u32 gl_bytes_per_pixel = CachedSurface::GetGLBytesPerPixel(format);
+    const auto& gpu = Core::System::GetInstance().GPU();
    if (morton_to_gl) {
        auto data = Tegra::Texture::UnswizzleTexture(
-            base, SurfaceParams::TextureFormatFromPixelFormat(format), stride, height,
+            *gpu.memory_manager->GpuToCpuAddress(base),
-            block_height);
+            SurfaceParams::TextureFormatFromPixelFormat(format), stride, height, block_height);
        std::memcpy(gl_buffer, data.data(), data.size());
    } else {
        // TODO(bunnei): Assumes the default rendering GOB size of 16 (128 lines). We should check
        // the configuration for this and perform more generic un/swizzle
-        LOG_WARNING(Render_OpenGL, "need to use correct swizzle/GOB parameters!");
+        NGLOG_WARNING(Render_OpenGL, "need to use correct swizzle/GOB parameters!");
-        VideoCore::MortonCopyPixels128(stride, height, bytes_per_pixel, gl_bytes_per_pixel,
+        VideoCore::MortonCopyPixels128(
-                                       Memory::GetPointer(base), gl_buffer, morton_to_gl);
+            stride, height, bytes_per_pixel, gl_bytes_per_pixel,
+            Memory::GetPointer(*gpu.memory_manager->GpuToCpuAddress(base)), gl_buffer,
+            morton_to_gl);
    }
 }
-static constexpr std::array<void (*)(u32, u32, u32, u8*, VAddr, VAddr, VAddr),
+static constexpr std::array<void (*)(u32, u32, u32, u8*, Tegra::GPUVAddr, Tegra::GPUVAddr,
+                                     Tegra::GPUVAddr),
                            SurfaceParams::MaxPixelFormat>
    morton_to_gl_fns = {
        MortonCopy<true, PixelFormat::ABGR8>,       MortonCopy<true, PixelFormat::B5G6R5>,
@@ -110,7 +111,8 @@ static constexpr std::array<void (*)(u32, u32, u32, u8*, VAddr, VAddr, VAddr),
        MortonCopy<true, PixelFormat::DXT23>,       MortonCopy<true, PixelFormat::DXT45>,
 };
-static constexpr std::array<void (*)(u32, u32, u32, u8*, VAddr, VAddr, VAddr),
+static constexpr std::array<void (*)(u32, u32, u32, u8*, Tegra::GPUVAddr, Tegra::GPUVAddr,
+                                     Tegra::GPUVAddr),
                            SurfaceParams::MaxPixelFormat>
    gl_to_morton_fns = {
        MortonCopy<false, PixelFormat::ABGR8>,
@@ -219,9 +221,9 @@ SurfaceParams SurfaceParams::FromInterval(SurfaceInterval interval) const {
    SurfaceParams params = *this;
    const u32 tiled_size = is_tiled ? 8 : 1;
    const u64 stride_tiled_bytes = BytesInPixels(stride * tiled_size);
-    VAddr aligned_start =
+    Tegra::GPUVAddr aligned_start =
        addr + Common::AlignDown(boost::icl::first(interval) - addr, stride_tiled_bytes);
-    VAddr aligned_end =
+    Tegra::GPUVAddr aligned_end =
        addr + Common::AlignUp(boost::icl::last_next(interval) - addr, stride_tiled_bytes);
    if (aligned_end - aligned_start > stride_tiled_bytes) {
@@ -342,6 +344,13 @@ bool SurfaceParams::CanTexCopy(const SurfaceParams& texcopy_params) const {
    return FromInterval(texcopy_params.GetInterval()).GetInterval() == texcopy_params.GetInterval();
 }
+VAddr SurfaceParams::GetCpuAddr() const {
+    // When this function is used, only cpu_addr or (GPU) addr should be set, not both
+    ASSERT(!(cpu_addr && addr));
+    const auto& gpu = Core::System::GetInstance().GPU();
+    return cpu_addr.get_value_or(*gpu.memory_manager->GpuToCpuAddress(addr));
+}
 bool CachedSurface::CanFill(const SurfaceParams& dest_surface,
                            SurfaceInterval fill_interval) const {
    if (type == SurfaceType::Fill && IsRegionValid(fill_interval) &&
@@ -349,9 +358,9 @@ bool CachedSurface::CanFill(const SurfaceParams& dest_surface,
        boost::icl::last_next(fill_interval) <= end && // dest_surface is within our fill range
        dest_surface.FromInterval(fill_interval).GetInterval() ==
            fill_interval) { // make sure interval is a rectangle in dest surface
-        if (fill_size * 8 != dest_surface.GetFormatBpp()) {
+        if (fill_size * CHAR_BIT != dest_surface.GetFormatBpp()) {
            // Check if bits repeat for our fill_size
-            const u32 dest_bytes_per_pixel = std::max(dest_surface.GetFormatBpp() / 8, 1u);
+            const u32 dest_bytes_per_pixel = std::max(dest_surface.GetFormatBpp() / CHAR_BIT, 1u);
            std::vector<u8> fill_test(fill_size * dest_bytes_per_pixel);
            for (u32 i = 0; i < dest_bytes_per_pixel; ++i)
@@ -456,15 +465,15 @@ void RasterizerCacheOpenGL::CopySurface(const Surface& src_surface, const Surfac
 }
 MICROPROFILE_DEFINE(OpenGL_SurfaceLoad, "OpenGL", "Surface Load", MP_RGB(128, 64, 192));
-void CachedSurface::LoadGLBuffer(VAddr load_start, VAddr load_end) {
+void CachedSurface::LoadGLBuffer(Tegra::GPUVAddr load_start, Tegra::GPUVAddr load_end) {
    ASSERT(type != SurfaceType::Fill);
-    u8* const texture_src_data = Memory::GetPointer(addr);
+    u8* const texture_src_data = Memory::GetPointer(GetCpuAddr());
    if (texture_src_data == nullptr)
        return;
    if (gl_buffer == nullptr) {
-        gl_buffer_size = width * height * GetGLBytesPerPixel(pixel_format);
+        gl_buffer_size = GetActualWidth() * GetActualHeight() * GetGLBytesPerPixel(pixel_format);
        gl_buffer.reset(new u8[gl_buffer_size]);
    }
@@ -479,14 +488,15 @@ void CachedSurface::LoadGLBuffer(VAddr load_start, VAddr load_end) {
        std::memcpy(&gl_buffer[start_offset], texture_src_data + start_offset,
                    bytes_per_pixel * width * height);
    } else {
-        morton_to_gl_fns[static_cast<size_t>(pixel_format)](
+        morton_to_gl_fns[static_cast<size_t>(pixel_format)](GetActualWidth(), block_height,
-            stride, block_height, height, &gl_buffer[0], addr, load_start, load_end);
+                                                            GetActualHeight(), &gl_buffer[0], addr,
+                                                            load_start, load_end);
    }
 }
 MICROPROFILE_DEFINE(OpenGL_SurfaceFlush, "OpenGL", "Surface Flush", MP_RGB(128, 192, 64));
-void CachedSurface::FlushGLBuffer(VAddr flush_start, VAddr flush_end) {
+void CachedSurface::FlushGLBuffer(Tegra::GPUVAddr flush_start, Tegra::GPUVAddr flush_end) {
-    u8* const dst_buffer = Memory::GetPointer(addr);
+    u8* const dst_buffer = Memory::GetPointer(GetCpuAddr());
    if (dst_buffer == nullptr)
        return;
@@ -536,7 +546,8 @@ void CachedSurface::UploadGLTexture(const MathUtil::Rectangle<u32>& rect, GLuint
    MICROPROFILE_SCOPE(OpenGL_TextureUL);
-    ASSERT(gl_buffer_size == width * height * GetGLBytesPerPixel(pixel_format));
+    ASSERT(gl_buffer_size ==
+           GetActualWidth() * GetActualHeight() * GetGLBytesPerPixel(pixel_format));
    // Load data from memory to the surface
    GLint x0 = static_cast<GLint>(rect.left);
@@ -571,11 +582,9 @@ void CachedSurface::UploadGLTexture(const MathUtil::Rectangle<u32>& rect, GLuint
    glActiveTexture(GL_TEXTURE0);
    if (tuple.compressed) {
        glCompressedTexImage2D(GL_TEXTURE_2D, 0, tuple.internal_format,
-                               static_cast<GLsizei>(rect.GetWidth()),
+                               static_cast<GLsizei>(rect.GetWidth() * GetCompresssionFactor()),
-                               static_cast<GLsizei>(rect.GetHeight()), 0,
+                               static_cast<GLsizei>(rect.GetHeight() * GetCompresssionFactor()), 0,
-                               rect.GetWidth() * rect.GetHeight() *
+                               size, &gl_buffer[buffer_offset]);
-                                   GetGLBytesPerPixel(pixel_format) / tuple.compression_factor,
-                               &gl_buffer[buffer_offset]);
    } else {
        glTexSubImage2D(GL_TEXTURE_2D, 0, x0, y0, static_cast<GLsizei>(rect.GetWidth()),
                        static_cast<GLsizei>(rect.GetHeight()), tuple.format, tuple.type,
@@ -945,6 +954,33 @@ Surface RasterizerCacheOpenGL::GetSurface(const SurfaceParams& params, ScaleMatc
    return surface;
 }
+boost::optional<Tegra::GPUVAddr> RasterizerCacheOpenGL::TryFindFramebufferGpuAddress(
+    VAddr cpu_addr) const {
+    // Tries to find the GPU address of a framebuffer based on the CPU address. This is because
+    // final output framebuffers are specified by CPU address, but internally our GPU cache uses GPU
+    // addresses. We iterate through all cached framebuffers, and compare their starting CPU address
+    // to the one provided. This is obviously not great, and won't work if the framebuffer overlaps
+    // surfaces.
+    std::vector<Tegra::GPUVAddr> gpu_addresses;
+    for (const auto& pair : surface_cache) {
+        for (const auto& surface : pair.second) {
+            const VAddr surface_cpu_addr = surface->GetCpuAddr();
+            if (cpu_addr >= surface_cpu_addr && cpu_addr < (surface_cpu_addr + surface->size)) {
+                ASSERT_MSG(cpu_addr == surface_cpu_addr, "overlapping surfaces are unsupported");
+                gpu_addresses.push_back(surface->addr);
+            }
+        }
+    }
+    if (gpu_addresses.empty()) {
+        return {};
+    }
+    ASSERT_MSG(gpu_addresses.size() == 1, ">1 surface is unsupported");
+    return gpu_addresses[0];
+}
 SurfaceRect_Tuple RasterizerCacheOpenGL::GetSurfaceSubRect(const SurfaceParams& params,
                                                           ScaleMatch match_res_scale,
                                                           bool load_if_create) {
@@ -1028,11 +1064,11 @@ Surface RasterizerCacheOpenGL::GetTextureSurface(const Tegra::Texture::FullTextu
    auto& gpu = Core::System::GetInstance().GPU();
    SurfaceParams params;
-    params.addr = gpu.memory_manager->PhysicalToVirtualAddress(config.tic.Address());
+    params.addr = config.tic.Address();
-    params.width = config.tic.Width();
-    params.height = config.tic.Height();
    params.is_tiled = config.tic.IsTiled();
    params.pixel_format = SurfaceParams::PixelFormatFromTextureFormat(config.tic.format);
+    params.width = config.tic.Width() / params.GetCompresssionFactor();
+    params.height = config.tic.Height() / params.GetCompresssionFactor();
    // TODO(Subv): Different types per component are not supported.
    ASSERT(config.tic.r_type.Value() == config.tic.g_type.Value() &&
@@ -1045,7 +1081,7 @@ Surface RasterizerCacheOpenGL::GetTextureSurface(const Tegra::Texture::FullTextu
        params.block_height = config.tic.BlockHeight();
    } else {
        // Use the texture-provided stride value if the texture isn't tiled.
-        params.stride = params.PixelsInBytes(config.tic.Pitch());
+        params.stride = static_cast<u32>(params.PixelsInBytes(config.tic.Pitch()));
    }
    params.UpdateParams();
@@ -1073,11 +1109,10 @@ Surface RasterizerCacheOpenGL::GetTextureSurface(const Tegra::Texture::FullTextu
 SurfaceSurfaceRect_Tuple RasterizerCacheOpenGL::GetFramebufferSurfaces(
    bool using_color_fb, bool using_depth_fb, const MathUtil::Rectangle<s32>& viewport) {
    const auto& regs = Core::System().GetInstance().GPU().Maxwell3D().regs;
-    const auto& memory_manager = Core::System().GetInstance().GPU().memory_manager;
    const auto& config = regs.rt[0];
    // TODO(bunnei): This is hard corded to use just the first render buffer
-    LOG_WARNING(Render_OpenGL, "hard-coded for render target 0!");
+    NGLOG_WARNING(Render_OpenGL, "hard-coded for render target 0!");
    // update resolution_scale_factor and reset cache if changed
    // TODO (bunnei): This code was ported as-is from Citra, and is technically not thread-safe. We
@@ -1106,7 +1141,7 @@ SurfaceSurfaceRect_Tuple RasterizerCacheOpenGL::GetFramebufferSurfaces(
    color_params.block_height = Tegra::Texture::TICEntry::DefaultBlockHeight;
    SurfaceParams depth_params = color_params;
-    color_params.addr = memory_manager->PhysicalToVirtualAddress(config.Address());
+    color_params.addr = config.Address();
    color_params.pixel_format = SurfaceParams::PixelFormatFromRenderTargetFormat(config.format);
    color_params.component_type = SurfaceParams::ComponentTypeFromRenderTarget(config.format);
    color_params.UpdateParams();
@@ -1122,8 +1157,8 @@ SurfaceSurfaceRect_Tuple RasterizerCacheOpenGL::GetFramebufferSurfaces(
    // Make sure that framebuffers don't overlap if both color and depth are being used
    if (using_color_fb && using_depth_fb &&
        boost::icl::length(color_vp_interval & depth_vp_interval)) {
-        LOG_CRITICAL(Render_OpenGL, "Color and depth framebuffer memory regions overlap; "
+        NGLOG_CRITICAL(Render_OpenGL, "Color and depth framebuffer memory regions overlap; "
-                                    "overlapping framebuffers not supported!");
+                                      "overlapping framebuffers not supported!");
        using_depth_fb = false;
    }
@@ -1222,7 +1257,8 @@ void RasterizerCacheOpenGL::DuplicateSurface(const Surface& src_surface,
    }
 }
-void RasterizerCacheOpenGL::ValidateSurface(const Surface& surface, VAddr addr, u64 size) {
+void RasterizerCacheOpenGL::ValidateSurface(const Surface& surface, Tegra::GPUVAddr addr,
+                                            u64 size) {
    if (size == 0)
        return;
@@ -1261,7 +1297,7 @@ void RasterizerCacheOpenGL::ValidateSurface(const Surface& surface, VAddr addr,
    }
 }
-void RasterizerCacheOpenGL::FlushRegion(VAddr addr, u64 size, Surface flush_surface) {
+void RasterizerCacheOpenGL::FlushRegion(Tegra::GPUVAddr addr, u64 size, Surface flush_surface) {
    if (size == 0)
        return;
@@ -1297,7 +1333,8 @@ void RasterizerCacheOpenGL::FlushAll() {
    FlushRegion(0, Kernel::VMManager::MAX_ADDRESS);
 }
-void RasterizerCacheOpenGL::InvalidateRegion(VAddr addr, u64 size, const Surface& region_owner) {
+void RasterizerCacheOpenGL::InvalidateRegion(Tegra::GPUVAddr addr, u64 size,
+                                             const Surface& region_owner) {
    if (size == 0)
        return;
@@ -1390,10 +1427,10 @@ void RasterizerCacheOpenGL::UnregisterSurface(const Surface& surface) {
    surface_cache.subtract({surface->GetInterval(), SurfaceSet{surface}});
 }
-void RasterizerCacheOpenGL::UpdatePagesCachedCount(VAddr addr, u64 size, int delta) {
+void RasterizerCacheOpenGL::UpdatePagesCachedCount(Tegra::GPUVAddr addr, u64 size, int delta) {
-    const u64 num_pages =
+    const u64 num_pages = ((addr + size - 1) >> Tegra::MemoryManager::PAGE_BITS) -
-        ((addr + size - 1) >> Memory::PAGE_BITS) - (addr >> Memory::PAGE_BITS) + 1;
+                          (addr >> Tegra::MemoryManager::PAGE_BITS) + 1;
-    const u64 page_start = addr >> Memory::PAGE_BITS;
+    const u64 page_start = addr >> Tegra::MemoryManager::PAGE_BITS;
    const u64 page_end = page_start + num_pages;
    // Interval maps will erase segments if count reaches 0, so if delta is negative we have to
@@ -1406,8 +1443,10 @@ void RasterizerCacheOpenGL::UpdatePagesCachedCount(VAddr addr, u64 size, int del
        const auto interval = pair.first & pages_interval;
        const int count = pair.second;
-        const VAddr interval_start_addr = boost::icl::first(interval) << Memory::PAGE_BITS;
+        const Tegra::GPUVAddr interval_start_addr = boost::icl::first(interval)
-        const VAddr interval_end_addr = boost::icl::last_next(interval) << Memory::PAGE_BITS;
+                                                    << Tegra::MemoryManager::PAGE_BITS;
+        const Tegra::GPUVAddr interval_end_addr = boost::icl::last_next(interval)
+                                                  << Tegra::MemoryManager::PAGE_BITS;
        const u64 interval_size = interval_end_addr - interval_start_addr;
        if (delta > 0 && count == delta)
diff --git a/src/video_core/renderer_opengl/gl_rasterizer_cache.h b/src/video_core/renderer_opengl/gl_rasterizer_cache.h
index e4cb3390f..55f1bdee8 100644
--- a/src/video_core/renderer_opengl/gl_rasterizer_cache.h
+++ b/src/video_core/renderer_opengl/gl_rasterizer_cache.h
@@ -17,12 +17,14 @@
 #ifdef __GNUC__
 #pragma GCC diagnostic pop
 #endif
+#include <boost/optional.hpp>
 #include <glad/glad.h>
 #include "common/assert.h"
 #include "common/common_funcs.h"
 #include "common/common_types.h"
 #include "common/math_util.h"
 #include "video_core/gpu.h"
+#include "video_core/memory_manager.h"
 #include "video_core/renderer_opengl/gl_resource_manager.h"
 #include "video_core/textures/texture.h"
@@ -30,9 +32,9 @@ struct CachedSurface;
 using Surface = std::shared_ptr<CachedSurface>;
 using SurfaceSet = std::set<Surface>;
-using SurfaceRegions = boost::icl::interval_set<VAddr>;
+using SurfaceRegions = boost::icl::interval_set<Tegra::GPUVAddr>;
-using SurfaceMap = boost::icl::interval_map<VAddr, Surface>;
+using SurfaceMap = boost::icl::interval_map<Tegra::GPUVAddr, Surface>;
-using SurfaceCache = boost::icl::interval_map<VAddr, SurfaceSet>;
+using SurfaceCache = boost::icl::interval_map<Tegra::GPUVAddr, SurfaceSet>;
 using SurfaceInterval = SurfaceCache::interval_type;
 static_assert(std::is_same<SurfaceRegions::interval_type, SurfaceCache::interval_type>() &&
@@ -82,23 +84,49 @@ struct SurfaceParams {
        Invalid = 4,
    };
-    static constexpr unsigned int GetFormatBpp(PixelFormat format) {
+    /**
+     * Gets the compression factor for the specified PixelFormat. This applies to just the
+     * "compressed width" and "compressed height", not the overall compression factor of a
+     * compressed image. This is used for maintaining proper surface sizes for compressed texture
+     * formats.
+     */
+    static constexpr u32 GetCompresssionFactor(PixelFormat format) {
        if (format == PixelFormat::Invalid)
            return 0;
-        constexpr std::array<unsigned int, MaxPixelFormat> bpp_table = {
+        constexpr std::array<u32, MaxPixelFormat> compression_factor_table = {{
+            1, // ABGR8
+            1, // B5G6R5
+            1, // A2B10G10R10
+            4, // DXT1
+            4, // DXT23
+            4, // DXT45
+        }};
+        ASSERT(static_cast<size_t>(format) < compression_factor_table.size());
+        return compression_factor_table[static_cast<size_t>(format)];
+    }
+    u32 GetCompresssionFactor() const {
+        return GetCompresssionFactor(pixel_format);
+    }
+    static constexpr u32 GetFormatBpp(PixelFormat format) {
+        if (format == PixelFormat::Invalid)
+            return 0;
+        constexpr std::array<u32, MaxPixelFormat> bpp_table = {{
            32,  // ABGR8
            16,  // B5G6R5
            32,  // A2B10G10R10
            64,  // DXT1
            128, // DXT23
            128, // DXT45
-        };
+        }};
        ASSERT(static_cast<size_t>(format) < bpp_table.size());
        return bpp_table[static_cast<size_t>(format)];
    }
-    unsigned int GetFormatBpp() const {
+    u32 GetFormatBpp() const {
        return GetFormatBpp(pixel_format);
    }
@@ -106,6 +134,8 @@ struct SurfaceParams {
        switch (format) {
        case Tegra::RenderTargetFormat::RGBA8_UNORM:
            return PixelFormat::ABGR8;
+        case Tegra::RenderTargetFormat::RGB10_A2_UNORM:
+            return PixelFormat::A2B10G10R10;
        default:
            NGLOG_CRITICAL(HW_GPU, "Unimplemented format={}", static_cast<u32>(format));
            UNREACHABLE();
@@ -251,6 +281,24 @@ struct SurfaceParams {
    // Returns the region of the biggest valid rectange within interval
    SurfaceInterval GetCopyableInterval(const Surface& src_surface) const;
+    /**
+     * Gets the actual width (in pixels) of the surface. This is provided because `width` is used
+     * for tracking the surface region in memory, which may be compressed for certain formats. In
+     * this scenario, `width` is actually the compressed width.
+     */
+    u32 GetActualWidth() const {
+        return width * GetCompresssionFactor();
+    }
+    /**
+     * Gets the actual height (in pixels) of the surface. This is provided because `height` is used
+     * for tracking the surface region in memory, which may be compressed for certain formats. In
+     * this scenario, `height` is actually the compressed height.
+     */
+    u32 GetActualHeight() const {
+        return height * GetCompresssionFactor();
+    }
    u32 GetScaledWidth() const {
        return width * res_scale;
    }
@@ -275,6 +323,8 @@ struct SurfaceParams {
        return pixels * GetFormatBpp(pixel_format) / CHAR_BIT;
    }
+    VAddr GetCpuAddr() const;
    bool ExactMatch(const SurfaceParams& other_surface) const;
    bool CanSubRect(const SurfaceParams& sub_surface) const;
    bool CanExpand(const SurfaceParams& expanded_surface) const;
@@ -283,8 +333,9 @@ struct SurfaceParams {
    MathUtil::Rectangle<u32> GetSubRect(const SurfaceParams& sub_surface) const;
    MathUtil::Rectangle<u32> GetScaledSubRect(const SurfaceParams& sub_surface) const;
-    VAddr addr = 0;
+    Tegra::GPUVAddr addr = 0;
-    VAddr end = 0;
+    Tegra::GPUVAddr end = 0;
+    boost::optional<VAddr> cpu_addr;
    u64 size = 0;
    u32 width = 0;
@@ -323,15 +374,15 @@ struct CachedSurface : SurfaceParams {
        if (format == PixelFormat::Invalid)
            return 0;
-        return SurfaceParams::GetFormatBpp(format) / 8;
+        return SurfaceParams::GetFormatBpp(format) / CHAR_BIT;
    }
    std::unique_ptr<u8[]> gl_buffer;
    size_t gl_buffer_size = 0;
    // Read/Write data in Switch memory to/from gl_buffer
-    void LoadGLBuffer(VAddr load_start, VAddr load_end);
+    void LoadGLBuffer(Tegra::GPUVAddr load_start, Tegra::GPUVAddr load_end);
-    void FlushGLBuffer(VAddr flush_start, VAddr flush_end);
+    void FlushGLBuffer(Tegra::GPUVAddr flush_start, Tegra::GPUVAddr flush_end);
    // Upload/Download data in gl_buffer in/to this surface's texture
    void UploadGLTexture(const MathUtil::Rectangle<u32>& rect, GLuint read_fb_handle,
@@ -360,6 +411,9 @@ public:
    Surface GetSurface(const SurfaceParams& params, ScaleMatch match_res_scale,
                       bool load_if_create);
+    /// Tries to find a framebuffer GPU address based on the provided CPU address
+    boost::optional<Tegra::GPUVAddr> TryFindFramebufferGpuAddress(VAddr cpu_addr) const;
    /// Attempt to find a subrect (resolution scaled) of a surface, otherwise loads a texture from
    /// Switch memory to OpenGL and caches it (if not already cached)
    SurfaceRect_Tuple GetSurfaceSubRect(const SurfaceParams& params, ScaleMatch match_res_scale,
@@ -379,10 +433,10 @@ public:
    SurfaceRect_Tuple GetTexCopySurface(const SurfaceParams& params);
    /// Write any cached resources overlapping the region back to memory (if dirty)
-    void FlushRegion(VAddr addr, u64 size, Surface flush_surface = nullptr);
+    void FlushRegion(Tegra::GPUVAddr addr, u64 size, Surface flush_surface = nullptr);
    /// Mark region as being invalidated by region_owner (nullptr if Switch memory)
-    void InvalidateRegion(VAddr addr, u64 size, const Surface& region_owner);
+    void InvalidateRegion(Tegra::GPUVAddr addr, u64 size, const Surface& region_owner);
    /// Flush all cached resources tracked by this cache manager
    void FlushAll();
@@ -391,7 +445,7 @@ private:
    void DuplicateSurface(const Surface& src_surface, const Surface& dest_surface);
    /// Update surface's texture for given region when necessary
-    void ValidateSurface(const Surface& surface, VAddr addr, u64 size);
+    void ValidateSurface(const Surface& surface, Tegra::GPUVAddr addr, u64 size);
    /// Create a new surface
    Surface CreateSurface(const SurfaceParams& params);
@@ -403,7 +457,7 @@ private:
    void UnregisterSurface(const Surface& surface);
    /// Increase/decrease the number of surface in pages touching the specified region
-    void UpdatePagesCachedCount(VAddr addr, u64 size, int delta);
+    void UpdatePagesCachedCount(Tegra::GPUVAddr addr, u64 size, int delta);
    SurfaceCache surface_cache;
    PageMap cached_pages;
diff --git a/src/video_core/renderer_opengl/gl_shader_decompiler.cpp b/src/video_core/renderer_opengl/gl_shader_decompiler.cpp
index 086424395..3dffb205d 100644
--- a/src/video_core/renderer_opengl/gl_shader_decompiler.cpp
+++ b/src/video_core/renderer_opengl/gl_shader_decompiler.cpp
@@ -519,7 +519,7 @@ private:
            }
            break;
        }
-        case OpCode::Type::FloatPredicate: {
+        case OpCode::Type::FloatSetPredicate: {
            std::string op_a = instr.fsetp.neg_a ? "-" : "";
            op_a += GetRegister(instr.gpr8);
@@ -570,6 +570,59 @@ private:
            }
            break;
        }
+        case OpCode::Type::FloatSet: {
+            std::string dest = GetRegister(instr.gpr0);
+            std::string op_a = instr.fset.neg_a ? "-" : "";
+            op_a += GetRegister(instr.gpr8);
+            if (instr.fset.abs_a) {
+                op_a = "abs(" + op_a + ')';
+            }
+            std::string op_b = instr.fset.neg_b ? "-" : "";
+            if (instr.is_b_imm) {
+                std::string imm = GetImmediate19(instr);
+                if (instr.fset.neg_imm)
+                    op_b += "(-" + imm + ')';
+                else
+                    op_b += imm;
+            } else {
+                if (instr.is_b_gpr) {
+                    op_b += GetRegister(instr.gpr20);
+                } else {
+                    op_b += GetUniform(instr.uniform);
+                }
+            }
+            if (instr.fset.abs_b) {
+                op_b = "abs(" + op_b + ")";
+            }
+            using Tegra::Shader::Pred;
+            ASSERT_MSG(instr.fset.pred39 == static_cast<u64>(Pred::UnusedIndex),
+                       "Compound predicates are not implemented");
+            // The fset instruction sets a register to 1.0 if the condition is true, and to 0
+            // otherwise.
+            using Tegra::Shader::PredCondition;
+            switch (instr.fset.cond) {
+            case PredCondition::LessThan:
+                SetDest(0, dest, "((" + op_a + ") < (" + op_b + ")) ? 1.0 : 0", 1, 1);
+                break;
+            case PredCondition::Equal:
+                SetDest(0, dest, "((" + op_a + ") == (" + op_b + ")) ? 1.0 : 0", 1, 1);
+                break;
+            case PredCondition::GreaterThan:
+                SetDest(0, dest, "((" + op_a + ") > (" + op_b + ")) ? 1.0 : 0", 1, 1);
+                break;
+            default:
+                NGLOG_CRITICAL(HW_GPU, "Unhandled predicate condition: {} (a: {}, b: {})",
+                               static_cast<unsigned>(instr.fset.cond.Value()), op_a, op_b);
+                UNREACHABLE();
+            }
+            break;
+        }
        default: {
            switch (opcode->GetId()) {
            case OpCode::Id::EXIT: {
diff --git a/src/video_core/renderer_opengl/renderer_opengl.cpp b/src/video_core/renderer_opengl/renderer_opengl.cpp
index ab0acb20a..77d1692f4 100644
--- a/src/video_core/renderer_opengl/renderer_opengl.cpp
+++ b/src/video_core/renderer_opengl/renderer_opengl.cpp
@@ -152,7 +152,8 @@ void RendererOpenGL::LoadFBToScreenInfo(const Tegra::FramebufferConfig& framebuf
        screen_info.display_texture = screen_info.texture.resource.handle;
        screen_info.display_texcoords = MathUtil::Rectangle<float>(0.f, 0.f, 1.f, 1.f);
-        Rasterizer()->FlushRegion(framebuffer_addr, size_in_bytes);
+        Memory::RasterizerFlushVirtualRegion(framebuffer_addr, size_in_bytes,
+                                             Memory::FlushMode::Flush);
        VideoCore::MortonCopyPixels128(framebuffer.width, framebuffer.height, bytes_per_pixel, 4,
                                       Memory::GetPointer(framebuffer_addr),
@@ -269,10 +270,9 @@ void RendererOpenGL::ConfigureFramebufferTexture(TextureInfo& texture,
    GLint internal_format;
    switch (framebuffer.pixel_format) {
    case Tegra::FramebufferConfig::PixelFormat::ABGR8:
-        // Use RGBA8 and swap in the fragment shader
        internal_format = GL_RGBA;
        texture.gl_format = GL_RGBA;
-        texture.gl_type = GL_UNSIGNED_INT_8_8_8_8;
+        texture.gl_type = GL_UNSIGNED_INT_8_8_8_8_REV;
        gl_framebuffer_data.resize(texture.width * texture.height * 4);
        break;
    default:
@@ -295,17 +295,18 @@ void RendererOpenGL::DrawScreenTriangles(const ScreenInfo& screen_info, float x,
    const auto& texcoords = screen_info.display_texcoords;
    auto left = texcoords.left;
    auto right = texcoords.right;
-    if (framebuffer_transform_flags != Tegra::FramebufferConfig::TransformFlags::Unset)
+    if (framebuffer_transform_flags != Tegra::FramebufferConfig::TransformFlags::Unset) {
        if (framebuffer_transform_flags == Tegra::FramebufferConfig::TransformFlags::FlipV) {
            // Flip the framebuffer vertically
            left = texcoords.right;
            right = texcoords.left;
        } else {
            // Other transformations are unsupported
-            LOG_CRITICAL(Render_OpenGL, "Unsupported framebuffer_transform_flags=%d",
+            NGLOG_CRITICAL(Render_OpenGL, "Unsupported framebuffer_transform_flags={}",
-                         framebuffer_transform_flags);
+                           static_cast<u32>(framebuffer_transform_flags));
            UNIMPLEMENTED();
        }
+    }
    std::array<ScreenRectVertex, 4> vertices = {{
        ScreenRectVertex(x, y, texcoords.top, left),
@@ -427,9 +428,9 @@ bool RendererOpenGL::Init() {
    const char* gpu_vendor{reinterpret_cast<char const*>(glGetString(GL_VENDOR))};
    const char* gpu_model{reinterpret_cast<char const*>(glGetString(GL_RENDERER))};
-    LOG_INFO(Render_OpenGL, "GL_VERSION: %s", gl_version);
+    NGLOG_INFO(Render_OpenGL, "GL_VERSION: {}", gl_version);
-    LOG_INFO(Render_OpenGL, "GL_VENDOR: %s", gpu_vendor);
+    NGLOG_INFO(Render_OpenGL, "GL_VENDOR: {}", gpu_vendor);
-    LOG_INFO(Render_OpenGL, "GL_RENDERER: %s", gpu_model);
+    NGLOG_INFO(Render_OpenGL, "GL_RENDERER: {}", gpu_model);
    Core::Telemetry().AddField(Telemetry::FieldType::UserSystem, "GPU_Vendor", gpu_vendor);
    Core::Telemetry().AddField(Telemetry::FieldType::UserSystem, "GPU_Model", gpu_model);
diff --git a/src/video_core/textures/decoders.cpp b/src/video_core/textures/decoders.cpp
index e0509f0ce..8b39b2bdf 100644
--- a/src/video_core/textures/decoders.cpp
+++ b/src/video_core/textures/decoders.cpp
@@ -4,6 +4,7 @@
 #include <cstring>
 #include "common/assert.h"
+#include "core/memory.h"
 #include "video_core/textures/decoders.h"
 #include "video_core/textures/texture.h"
@@ -26,9 +27,8 @@ static u32 GetSwizzleOffset(u32 x, u32 y, u32 image_width, u32 bytes_per_pixel,
    return address;
 }
-static void CopySwizzledData(u32 width, u32 height, u32 bytes_per_pixel, u32 out_bytes_per_pixel,
+void CopySwizzledData(u32 width, u32 height, u32 bytes_per_pixel, u32 out_bytes_per_pixel,
-                             u8* swizzled_data, u8* unswizzled_data, bool unswizzle,
+                      u8* swizzled_data, u8* unswizzled_data, bool unswizzle, u32 block_height) {
-                             u32 block_height) {
    u8* data_ptrs[2];
    for (unsigned y = 0; y < height; ++y) {
        for (unsigned x = 0; x < width; ++x) {
diff --git a/src/video_core/textures/decoders.h b/src/video_core/textures/decoders.h
index a700911cf..2562c4b06 100644
--- a/src/video_core/textures/decoders.h
+++ b/src/video_core/textures/decoders.h
@@ -17,6 +17,10 @@ namespace Texture {
 std::vector<u8> UnswizzleTexture(VAddr address, TextureFormat format, u32 width, u32 height,
                                 u32 block_height = TICEntry::DefaultBlockHeight);
+/// Copies texture data from a buffer and performs swizzling/unswizzling as necessary.
+void CopySwizzledData(u32 width, u32 height, u32 bytes_per_pixel, u32 out_bytes_per_pixel,
+                      u8* swizzled_data, u8* unswizzled_data, bool unswizzle, u32 block_height);
 /**
 * Decodes an unswizzled texture into a A8R8G8B8 texture.
 */
diff --git a/src/video_core/video_core.cpp b/src/video_core/video_core.cpp
index 289140f31..89dc8ed1e 100644
--- a/src/video_core/video_core.cpp
+++ b/src/video_core/video_core.cpp
@@ -24,9 +24,9 @@ bool Init(EmuWindow* emu_window) {
    g_renderer = std::make_unique<RendererOpenGL>();
    g_renderer->SetWindow(g_emu_window);
    if (g_renderer->Init()) {
-        LOG_DEBUG(Render, "initialized OK");
+        NGLOG_DEBUG(Render, "initialized OK");
    } else {
-        LOG_CRITICAL(Render, "initialization failed !");
+        NGLOG_CRITICAL(Render, "initialization failed !");
        return false;
    }
    return true;
@@ -36,7 +36,7 @@ bool Init(EmuWindow* emu_window) {
 void Shutdown() {
    g_renderer.reset();
-    LOG_DEBUG(Render, "shutdown OK");
+    NGLOG_DEBUG(Render, "shutdown OK");
 }
 } // namespace VideoCore
diff --git a/src/yuzu/debugger/graphics/graphics_surface.cpp b/src/yuzu/debugger/graphics/graphics_surface.cpp
index 1e4844b57..1fbca8ad0 100644
--- a/src/yuzu/debugger/graphics/graphics_surface.cpp
+++ b/src/yuzu/debugger/graphics/graphics_surface.cpp
@@ -25,6 +25,8 @@ static Tegra::Texture::TextureFormat ConvertToTextureFormat(
    switch (render_target_format) {
    case Tegra::RenderTargetFormat::RGBA8_UNORM:
        return Tegra::Texture::TextureFormat::A8R8G8B8;
+    case Tegra::RenderTargetFormat::RGB10_A2_UNORM:
+        return Tegra::Texture::TextureFormat::A2B10G10R10;
    default:
        UNIMPLEMENTED_MSG("Unimplemented RT format");
    }
@@ -376,10 +378,10 @@ void GraphicsSurfaceWidget::OnUpdate() {
    // TODO: Implement a good way to visualize alpha components!
    QImage decoded_image(surface_width, surface_height, QImage::Format_ARGB32);
-    VAddr address = gpu.memory_manager->PhysicalToVirtualAddress(surface_address);
+    boost::optional<VAddr> address = gpu.memory_manager->GpuToCpuAddress(surface_address);
    auto unswizzled_data =
-        Tegra::Texture::UnswizzleTexture(address, surface_format, surface_width, surface_height);
+        Tegra::Texture::UnswizzleTexture(*address, surface_format, surface_width, surface_height);
    auto texture_data = Tegra::Texture::DecodeTexture(unswizzled_data, surface_format,
                                                      surface_width, surface_height);
@@ -435,9 +437,9 @@ void GraphicsSurfaceWidget::SaveSurface() {
            pixmap->save(&file, "PNG");
    } else if (selectedFilter == bin_filter) {
        auto& gpu = Core::System::GetInstance().GPU();
-        VAddr address = gpu.memory_manager->PhysicalToVirtualAddress(surface_address);
+        boost::optional<VAddr> address = gpu.memory_manager->GpuToCpuAddress(surface_address);
-        const u8* buffer = Memory::GetPointer(address);
+        const u8* buffer = Memory::GetPointer(*address);
        ASSERT_MSG(buffer != nullptr, "Memory not accessible");
        QFile file(filename);
diff --git a/src/yuzu/debugger/wait_tree.cpp b/src/yuzu/debugger/wait_tree.cpp
index cae2864e5..acc4c2e0b 100644
--- a/src/yuzu/debugger/wait_tree.cpp
+++ b/src/yuzu/debugger/wait_tree.cpp
@@ -6,8 +6,8 @@
 #include "yuzu/util/util.h"
 #include "core/core.h"
-#include "core/hle/kernel/condition_variable.h"
 #include "core/hle/kernel/event.h"
+#include "core/hle/kernel/handle_table.h"
 #include "core/hle/kernel/mutex.h"
 #include "core/hle/kernel/thread.h"
 #include "core/hle/kernel/timer.h"
@@ -67,6 +67,29 @@ QString WaitTreeText::GetText() const {
    return text;
 }
+WaitTreeMutexInfo::WaitTreeMutexInfo(VAddr mutex_address) : mutex_address(mutex_address) {
+    mutex_value = Memory::Read32(mutex_address);
+    owner_handle = static_cast<Kernel::Handle>(mutex_value & Kernel::Mutex::MutexOwnerMask);
+    owner = Kernel::g_handle_table.Get<Kernel::Thread>(owner_handle);
+}
+QString WaitTreeMutexInfo::GetText() const {
+    return tr("waiting for mutex 0x%1").arg(mutex_address, 16, 16, QLatin1Char('0'));
+}
+std::vector<std::unique_ptr<WaitTreeItem>> WaitTreeMutexInfo::GetChildren() const {
+    std::vector<std::unique_ptr<WaitTreeItem>> list;
+    bool has_waiters = (mutex_value & Kernel::Mutex::MutexHasWaitersFlag) != 0;
+    list.push_back(std::make_unique<WaitTreeText>(tr("has waiters: %1").arg(has_waiters)));
+    list.push_back(std::make_unique<WaitTreeText>(
+        tr("owner handle: 0x%1").arg(owner_handle, 8, 16, QLatin1Char('0'))));
+    if (owner != nullptr)
+        list.push_back(std::make_unique<WaitTreeThread>(*owner));
+    return list;
+}
 WaitTreeWaitObject::WaitTreeWaitObject(const Kernel::WaitObject& o) : object(o) {}
 bool WaitTreeExpandableItem::IsExpandable() const {
@@ -84,11 +107,6 @@ std::unique_ptr<WaitTreeWaitObject> WaitTreeWaitObject::make(const Kernel::WaitO
    switch (object.GetHandleType()) {
    case Kernel::HandleType::Event:
        return std::make_unique<WaitTreeEvent>(static_cast<const Kernel::Event&>(object));
-    case Kernel::HandleType::Mutex:
-        return std::make_unique<WaitTreeMutex>(static_cast<const Kernel::Mutex&>(object));
-    case Kernel::HandleType::ConditionVariable:
-        return std::make_unique<WaitTreeConditionVariable>(
-            static_cast<const Kernel::ConditionVariable&>(object));
    case Kernel::HandleType::Timer:
        return std::make_unique<WaitTreeTimer>(static_cast<const Kernel::Timer&>(object));
    case Kernel::HandleType::Thread:
@@ -160,6 +178,9 @@ QString WaitTreeThread::GetText() const {
    case THREADSTATUS_WAIT_SYNCH_ANY:
        status = tr("waiting for objects");
        break;
+    case THREADSTATUS_WAIT_MUTEX:
+        status = tr("waiting for mutex");
+        break;
    case THREADSTATUS_DORMANT:
        status = tr("dormant");
        break;
@@ -186,6 +207,7 @@ QColor WaitTreeThread::GetColor() const {
        return QColor(Qt::GlobalColor::darkYellow);
    case THREADSTATUS_WAIT_SYNCH_ALL:
    case THREADSTATUS_WAIT_SYNCH_ANY:
+    case THREADSTATUS_WAIT_MUTEX:
        return QColor(Qt::GlobalColor::red);
    case THREADSTATUS_DORMANT:
        return QColor(Qt::GlobalColor::darkCyan);
@@ -225,11 +247,11 @@ std::vector<std::unique_ptr<WaitTreeItem>> WaitTreeThread::GetChildren() const {
    list.push_back(std::make_unique<WaitTreeText>(
        tr("last running ticks = %1").arg(thread.last_running_ticks)));
-    if (thread.held_mutexes.empty()) {
+    if (thread.mutex_wait_address != 0)
-        list.push_back(std::make_unique<WaitTreeText>(tr("not holding mutex")));
+        list.push_back(std::make_unique<WaitTreeMutexInfo>(thread.mutex_wait_address));
-    } else {
+    else
-        list.push_back(std::make_unique<WaitTreeMutexList>(thread.held_mutexes));
+        list.push_back(std::make_unique<WaitTreeText>(tr("not waiting for mutex")));
-    }
    if (thread.status == THREADSTATUS_WAIT_SYNCH_ANY ||
        thread.status == THREADSTATUS_WAIT_SYNCH_ALL) {
        list.push_back(std::make_unique<WaitTreeObjectList>(thread.wait_objects,
@@ -250,33 +272,6 @@ std::vector<std::unique_ptr<WaitTreeItem>> WaitTreeEvent::GetChildren() const {
    return list;
 }
-WaitTreeMutex::WaitTreeMutex(const Kernel::Mutex& object) : WaitTreeWaitObject(object) {}
-std::vector<std::unique_ptr<WaitTreeItem>> WaitTreeMutex::GetChildren() const {
-    std::vector<std::unique_ptr<WaitTreeItem>> list(WaitTreeWaitObject::GetChildren());
-    const auto& mutex = static_cast<const Kernel::Mutex&>(object);
-    if (mutex.GetHasWaiters()) {
-        list.push_back(std::make_unique<WaitTreeText>(tr("locked by thread:")));
-        list.push_back(std::make_unique<WaitTreeThread>(*mutex.GetHoldingThread()));
-    } else {
-        list.push_back(std::make_unique<WaitTreeText>(tr("free")));
-    }
-    return list;
-}
-WaitTreeConditionVariable::WaitTreeConditionVariable(const Kernel::ConditionVariable& object)
-    : WaitTreeWaitObject(object) {}
-std::vector<std::unique_ptr<WaitTreeItem>> WaitTreeConditionVariable::GetChildren() const {
-    std::vector<std::unique_ptr<WaitTreeItem>> list(WaitTreeWaitObject::GetChildren());
-    const auto& condition_variable = static_cast<const Kernel::ConditionVariable&>(object);
-    list.push_back(std::make_unique<WaitTreeText>(
-        tr("available count = %1").arg(condition_variable.GetAvailableCount())));
-    return list;
-}
 WaitTreeTimer::WaitTreeTimer(const Kernel::Timer& object) : WaitTreeWaitObject(object) {}
 std::vector<std::unique_ptr<WaitTreeItem>> WaitTreeTimer::GetChildren() const {
@@ -293,21 +288,6 @@ std::vector<std::unique_ptr<WaitTreeItem>> WaitTreeTimer::GetChildren() const {
    return list;
 }
-WaitTreeMutexList::WaitTreeMutexList(
-    const boost::container::flat_set<Kernel::SharedPtr<Kernel::Mutex>>& list)
-    : mutex_list(list) {}
-QString WaitTreeMutexList::GetText() const {
-    return tr("holding mutexes");
-}
-std::vector<std::unique_ptr<WaitTreeItem>> WaitTreeMutexList::GetChildren() const {
-    std::vector<std::unique_ptr<WaitTreeItem>> list(mutex_list.size());
-    std::transform(mutex_list.begin(), mutex_list.end(), list.begin(),
-                   [](const auto& t) { return std::make_unique<WaitTreeMutex>(*t); });
-    return list;
-}
 WaitTreeThreadList::WaitTreeThreadList(const std::vector<Kernel::SharedPtr<Kernel::Thread>>& list)
    : thread_list(list) {}
diff --git a/src/yuzu/debugger/wait_tree.h b/src/yuzu/debugger/wait_tree.h
index e538174eb..300ba9ae4 100644
--- a/src/yuzu/debugger/wait_tree.h
+++ b/src/yuzu/debugger/wait_tree.h
@@ -16,8 +16,6 @@ class EmuThread;
 namespace Kernel {
 class WaitObject;
 class Event;
-class Mutex;
-class ConditionVariable;
 class Thread;
 class Timer;
 } // namespace Kernel
@@ -61,6 +59,20 @@ public:
    bool IsExpandable() const override;
 };
+class WaitTreeMutexInfo : public WaitTreeExpandableItem {
+    Q_OBJECT
+public:
+    explicit WaitTreeMutexInfo(VAddr mutex_address);
+    QString GetText() const override;
+    std::vector<std::unique_ptr<WaitTreeItem>> GetChildren() const override;
+private:
+    VAddr mutex_address;
+    u32 mutex_value;
+    Kernel::Handle owner_handle;
+    Kernel::SharedPtr<Kernel::Thread> owner;
+};
 class WaitTreeWaitObject : public WaitTreeExpandableItem {
    Q_OBJECT
 public:
@@ -104,20 +116,6 @@ public:
    std::vector<std::unique_ptr<WaitTreeItem>> GetChildren() const override;
 };
-class WaitTreeMutex : public WaitTreeWaitObject {
-    Q_OBJECT
-public:
-    explicit WaitTreeMutex(const Kernel::Mutex& object);
-    std::vector<std::unique_ptr<WaitTreeItem>> GetChildren() const override;
-};
-class WaitTreeConditionVariable : public WaitTreeWaitObject {
-    Q_OBJECT
-public:
-    explicit WaitTreeConditionVariable(const Kernel::ConditionVariable& object);
-    std::vector<std::unique_ptr<WaitTreeItem>> GetChildren() const override;
-};
 class WaitTreeTimer : public WaitTreeWaitObject {
    Q_OBJECT
 public:
@@ -125,19 +123,6 @@ public:
    std::vector<std::unique_ptr<WaitTreeItem>> GetChildren() const override;
 };
-class WaitTreeMutexList : public WaitTreeExpandableItem {
-    Q_OBJECT
-public:
-    explicit WaitTreeMutexList(
-        const boost::container::flat_set<Kernel::SharedPtr<Kernel::Mutex>>& list);
-    QString GetText() const override;
-    std::vector<std::unique_ptr<WaitTreeItem>> GetChildren() const override;
-private:
-    const boost::container::flat_set<Kernel::SharedPtr<Kernel::Mutex>>& mutex_list;
-};
 class WaitTreeThreadList : public WaitTreeExpandableItem {
    Q_OBJECT
 public: