65 files changed, 809 insertions, 338 deletions

diff --git a/.travis.descriptor.json.in b/.travis.descriptor.json.in
deleted file mode 100644
index fac2eee93..000000000
--- a/.travis.descriptor.json.in
+++ /dev/null
@@ -1,23 +0,0 @@
-{
-    "package": {
-        "name": "nightly",
-        "repo": "citra",
-        "subject": "citra-emu",
-        "desc": "Build created on @BUILD_DATE@",
-        "website_url": "www.citra-emu.org",
-        "public_download_numbers": false,
-        "public_stats": false
-    },
-    "version": {
-        "name": "nightly-@GIT_REV@",
-        "released": "@BUILD_DATE@",
-        "gpgSign": false
-    },
-    "files": [
-        {
-            "includePattern": "artifacts/(.*\\.tar.xz)",
-            "uploadPattern": "$1"
-        }
-    ],
-    "publish": true
-}
diff --git a/.travis.yml b/.travis.yml
index 414168997..cf1e1e26c 100644
--- a/.travis.yml
+++ b/.travis.yml
@@ -29,8 +29,11 @@ script: "./.travis-build.sh"
 after_success: "./.travis-upload.sh"
 
 deploy:
-  provider: bintray
-  file: "${HOME}/build/${TRAVIS_REPO_SLUG}/.travis.descriptor.json"
-  user: citra-emu
-  key:
-    secure: "Hdj+/UyDYbvOOXJdlrV0n//t6tAnUqMEy4+Ppb84aF+na4B2hoc+RQ6rfbVXnFE0YKAKJme0J00GVi4JcXViOil2P/wHWxniWBlpz1vHLPT15+p1qi8cPQ0yKMAq9+1kYe6APkHjjsZxHAR6cf1fckI/rPfx/lyuDQOpVq07pJg="
\ No newline at end of file
+  provider: releases
+  api_key:
+    secure: Mck15DIWaJdxDiS3aYVlM9N3G6y8VKUI1rnwII7/iolfm1s94U+tgvbheZDmT7SSbFyaGaYO/E8HrV/uZR9Vvs7ev20sHsTN1u60OTWfDIIyHs9SqjhcGbtq95m9/dMFschOYqTOR+gAs5BsxjuoeAotHdhpQEwvkO2oo5oR0zhGy45gjFnVvtcxT/IfpZBIpVgcK3aLb9zT6ekcJbSiPmEB15iLq3xXd0nFUNtEZdX3D6Veye4n5jB6n72qN8JVoKvPZAwaC2K0pZxpcGJaXDchLsw1q+4eCvdz6UJfUemeQ/uMAmjfeQ3wrzYGXe3nCM3WmX5wosCsB0mw4zYatzl3si6CZ1W+0GkV4Rwlx03dfp7v3EeFhTsXYCaXqhwuLZnWOLUik8t9vaSoFUx4nUIRwfO9kAMUJQSpLuHNO2nT01s3GxvqxzczuLQ9he5nGSi0RRodUzDwek1qUp6I4uV3gRHKz4B07YIc1i2fK88NLXjyQ0uLVZ+7Oq1+kgDp6+N7vvXXZ5qZ17tdaysSbKEE0Y8zsoXw7Rk1tPN19vrCS+TSpomNMyQyne1k+I5iZ/qkxPTLAS5qI6Utc2dL3GJdxWRAEfGNO9AIX3GV/jmmKfdcvwGsCYP8hxqs5vLYfgacw3D8NLf1941lQUwavC17jm9EV9g5G3Pn1Cp516E=
+  file_glob: true
+  file: "artifacts/*.tar.xz"
+  skip_cleanup: true
+  on:
+    repo: citra-emu/citra-nightly
\ No newline at end of file
diff --git a/CMakeLists.txt b/CMakeLists.txt
index e232d99a4..52a1fd492 100644
--- a/CMakeLists.txt
+++ b/CMakeLists.txt
@@ -253,11 +253,6 @@ get_git_head_revision(GIT_REF_SPEC GIT_REV)
 git_describe(GIT_DESC --always --long --dirty)
 git_branch_name(GIT_BRANCH)
 
-# configure the bintray travis release json
-string(TIMESTAMP BUILD_DATE "%Y-%m-%d")
-configure_file("${CMAKE_SOURCE_DIR}/.travis.descriptor.json.in" "${CMAKE_SOURCE_DIR}/.travis.descriptor.json" @ONLY)
-unset(BUILD_DATE)
-
 set(INI_PREFIX "${CMAKE_CURRENT_SOURCE_DIR}/externals/inih")
 include_directories(${INI_PREFIX})
 add_subdirectory(${INI_PREFIX})
diff --git a/appveyor.yml b/appveyor.yml
index e4e1045de..c07559479 100644
--- a/appveyor.yml
+++ b/appveyor.yml
@@ -42,7 +42,7 @@ after_build:
         # set the build names as env vars so the artifacts can upload them
         $env:MSVC_BUILD_NAME = $MSVC_BUILD_NAME
         $env:MSVC_BUILD_PDB = $MSVC_BUILD_PDB
-        $env:BINTRAY_VERSION = $BINTRAY_VERSION
+        $env:GITREV = $GITREV
 
         7z a -tzip $MSVC_BUILD_PDB .\build\bin\release\*.pdb
         rm .\build\bin\release\*.pdb
@@ -60,14 +60,17 @@ artifacts:
     type: zip
 
 deploy:
-- provider: BinTray
-  username: citra-emu
-  api_key:
-    secure: qaZZfXp5Vv9xw5EmCxKLhGnNYhiNMDCdP84l0+h6HetfmvFlvHxyohWEVokLjl7d
-  subject: citra-emu
-  repo: citra
-  package: nightly
-  version: $(BINTRAY_VERSION)
-  publish: true
+  provider: GitHub
+  release: nightly-$(appveyor_build_number)
+  description: |
+    Citra nightly releases. Please choose the correct download for your operating system from the list below.
+
+    Short Commit Hash $(GITREV)
+  auth_token:
+    secure: "dbpsMC/MgPKWFNJCXpQl4cR8FYhepkPLjgNp/pRMktZ8oLKTqPYErfreaIxb/4P1"
+  artifact: msvcbuild
+  draft: false
+  prerelease: false
   on:
     branch: master
+    appveyor_repo_name: citra-emu/citra-nightly
diff --git a/src/citra/config.cpp b/src/citra/config.cpp
index 98f093258..bd8ac563b 100644
--- a/src/citra/config.cpp
+++ b/src/citra/config.cpp
@@ -63,8 +63,7 @@ void Config::ReadValues() {
     // Renderer
     Settings::values.use_hw_renderer = sdl2_config->GetBoolean("Renderer", "use_hw_renderer", true);
     Settings::values.use_shader_jit = sdl2_config->GetBoolean("Renderer", "use_shader_jit", true);
-    Settings::values.use_scaled_resolution =
-        sdl2_config->GetBoolean("Renderer", "use_scaled_resolution", false);
+    Settings::values.resolution_factor = sdl2_config->GetReal("Renderer", "resolution_factor", 1.0);
     Settings::values.use_vsync = sdl2_config->GetBoolean("Renderer", "use_vsync", false);
     Settings::values.toggle_framelimit =
         sdl2_config->GetBoolean("Renderer", "toggle_framelimit", true);
diff --git a/src/citra/default_ini.h b/src/citra/default_ini.h
index bb4720d25..7996813b4 100644
--- a/src/citra/default_ini.h
+++ b/src/citra/default_ini.h
@@ -51,9 +51,10 @@ use_hw_renderer =
 # 0: Interpreter (slow), 1 (default): JIT (fast)
 use_shader_jit =
 
-# Whether to use native 3DS screen resolution or to scale rendering resolution to the displayed screen size.
-# 0 (default): Native, 1: Scaled
-use_scaled_resolution =
+# Resolution scale factor
+# 0: Auto (scales resolution to window size), 1: Native 3DS screen resolution, Otherwise a scale
+# factor for the 3DS resolution
+resolution_factor =
 
 # Whether to enable V-Sync (caps the framerate at 60FPS) or not.
 # 0 (default): Off, 1: On
diff --git a/src/citra/emu_window/emu_window_sdl2.cpp b/src/citra/emu_window/emu_window_sdl2.cpp
index b0d82b670..81a3abe3f 100644
--- a/src/citra/emu_window/emu_window_sdl2.cpp
+++ b/src/citra/emu_window/emu_window_sdl2.cpp
@@ -19,16 +19,22 @@
 
 void EmuWindow_SDL2::OnMouseMotion(s32 x, s32 y) {
     TouchMoved((unsigned)std::max(x, 0), (unsigned)std::max(y, 0));
+    motion_emu->Tilt(x, y);
 }
 
 void EmuWindow_SDL2::OnMouseButton(u32 button, u8 state, s32 x, s32 y) {
-    if (button != SDL_BUTTON_LEFT)
-        return;
-
-    if (state == SDL_PRESSED) {
-        TouchPressed((unsigned)std::max(x, 0), (unsigned)std::max(y, 0));
-    } else {
-        TouchReleased();
+    if (button == SDL_BUTTON_LEFT) {
+        if (state == SDL_PRESSED) {
+            TouchPressed((unsigned)std::max(x, 0), (unsigned)std::max(y, 0));
+        } else {
+            TouchReleased();
+        }
+    } else if (button == SDL_BUTTON_RIGHT) {
+        if (state == SDL_PRESSED) {
+            motion_emu->BeginTilt(x, y);
+        } else {
+            motion_emu->EndTilt();
+        }
     }
 }
 
@@ -54,6 +60,7 @@ EmuWindow_SDL2::EmuWindow_SDL2() {
     keyboard_id = KeyMap::NewDeviceId();
 
     ReloadSetKeymaps();
+    motion_emu = std::make_unique<Motion::MotionEmu>(*this);
 
     SDL_SetMainReady();
 
@@ -109,6 +116,7 @@ EmuWindow_SDL2::EmuWindow_SDL2() {
 EmuWindow_SDL2::~EmuWindow_SDL2() {
     SDL_GL_DeleteContext(gl_context);
     SDL_Quit();
+    motion_emu = nullptr;
 }
 
 void EmuWindow_SDL2::SwapBuffers() {
diff --git a/src/citra/emu_window/emu_window_sdl2.h b/src/citra/emu_window/emu_window_sdl2.h
index c8cd919c6..b1cbf16d7 100644
--- a/src/citra/emu_window/emu_window_sdl2.h
+++ b/src/citra/emu_window/emu_window_sdl2.h
@@ -4,8 +4,10 @@
 
 #pragma once
 
+#include <memory>
 #include <utility>
 #include "core/frontend/emu_window.h"
+#include "core/frontend/motion_emu.h"
 
 struct SDL_Window;
 
@@ -61,4 +63,7 @@ private:
 
     /// Device id of keyboard for use with KeyMap
     int keyboard_id;
+
+    /// Motion sensors emulation
+    std::unique_ptr<Motion::MotionEmu> motion_emu;
 };
diff --git a/src/citra_qt/bootmanager.cpp b/src/citra_qt/bootmanager.cpp
index 57fde6caa..948db384d 100644
--- a/src/citra_qt/bootmanager.cpp
+++ b/src/citra_qt/bootmanager.cpp
@@ -99,7 +99,7 @@ private:
 };
 
 GRenderWindow::GRenderWindow(QWidget* parent, EmuThread* emu_thread)
-    : QWidget(parent), keyboard_id(0), emu_thread(emu_thread), child(nullptr) {
+    : QWidget(parent), child(nullptr), keyboard_id(0), emu_thread(emu_thread) {
 
     std::string window_title =
         Common::StringFromFormat("Citra | %s-%s", Common::g_scm_branch, Common::g_scm_desc);
@@ -191,6 +191,7 @@ qreal GRenderWindow::windowPixelRatio() {
 }
 
 void GRenderWindow::closeEvent(QCloseEvent* event) {
+    motion_emu = nullptr;
     emit Closed();
     QWidget::closeEvent(event);
 }
@@ -204,11 +205,13 @@ void GRenderWindow::keyReleaseEvent(QKeyEvent* event) {
 }
 
 void GRenderWindow::mousePressEvent(QMouseEvent* event) {
+    auto pos = event->pos();
     if (event->button() == Qt::LeftButton) {
-        auto pos = event->pos();
         qreal pixelRatio = windowPixelRatio();
         this->TouchPressed(static_cast<unsigned>(pos.x() * pixelRatio),
                            static_cast<unsigned>(pos.y() * pixelRatio));
+    } else if (event->button() == Qt::RightButton) {
+        motion_emu->BeginTilt(pos.x(), pos.y());
     }
 }
 
@@ -217,11 +220,14 @@ void GRenderWindow::mouseMoveEvent(QMouseEvent* event) {
     qreal pixelRatio = windowPixelRatio();
     this->TouchMoved(std::max(static_cast<unsigned>(pos.x() * pixelRatio), 0u),
                      std::max(static_cast<unsigned>(pos.y() * pixelRatio), 0u));
+    motion_emu->Tilt(pos.x(), pos.y());
 }
 
 void GRenderWindow::mouseReleaseEvent(QMouseEvent* event) {
     if (event->button() == Qt::LeftButton)
         this->TouchReleased();
+    else if (event->button() == Qt::RightButton)
+        motion_emu->EndTilt();
 }
 
 void GRenderWindow::ReloadSetKeymaps() {
@@ -279,11 +285,13 @@ void GRenderWindow::OnMinimalClientAreaChangeRequest(
 }
 
 void GRenderWindow::OnEmulationStarting(EmuThread* emu_thread) {
+    motion_emu = std::make_unique<Motion::MotionEmu>(*this);
     this->emu_thread = emu_thread;
     child->DisablePainting();
 }
 
 void GRenderWindow::OnEmulationStopping() {
+    motion_emu = nullptr;
     emu_thread = nullptr;
     child->EnablePainting();
 }
diff --git a/src/citra_qt/bootmanager.h b/src/citra_qt/bootmanager.h
index 43015390b..7dac1c480 100644
--- a/src/citra_qt/bootmanager.h
+++ b/src/citra_qt/bootmanager.h
@@ -11,6 +11,7 @@
 #include <QThread>
 #include "common/thread.h"
 #include "core/frontend/emu_window.h"
+#include "core/frontend/motion_emu.h"
 
 class QKeyEvent;
 class QScreen;
@@ -156,6 +157,9 @@ private:
 
     EmuThread* emu_thread;
 
+    /// Motion sensors emulation
+    std::unique_ptr<Motion::MotionEmu> motion_emu;
+
 protected:
     void showEvent(QShowEvent* event) override;
 };
diff --git a/src/citra_qt/config.cpp b/src/citra_qt/config.cpp
index c904c4b00..8021667d0 100644
--- a/src/citra_qt/config.cpp
+++ b/src/citra_qt/config.cpp
@@ -44,8 +44,7 @@ void Config::ReadValues() {
     qt_config->beginGroup("Renderer");
     Settings::values.use_hw_renderer = qt_config->value("use_hw_renderer", true).toBool();
     Settings::values.use_shader_jit = qt_config->value("use_shader_jit", true).toBool();
-    Settings::values.use_scaled_resolution =
-        qt_config->value("use_scaled_resolution", false).toBool();
+    Settings::values.resolution_factor = qt_config->value("resolution_factor", 1.0).toFloat();
     Settings::values.use_vsync = qt_config->value("use_vsync", false).toBool();
     Settings::values.toggle_framelimit = qt_config->value("toggle_framelimit", true).toBool();
 
@@ -152,7 +151,7 @@ void Config::SaveValues() {
     qt_config->beginGroup("Renderer");
     qt_config->setValue("use_hw_renderer", Settings::values.use_hw_renderer);
     qt_config->setValue("use_shader_jit", Settings::values.use_shader_jit);
-    qt_config->setValue("use_scaled_resolution", Settings::values.use_scaled_resolution);
+    qt_config->setValue("resolution_factor", (double)Settings::values.resolution_factor);
     qt_config->setValue("use_vsync", Settings::values.use_vsync);
     qt_config->setValue("toggle_framelimit", Settings::values.toggle_framelimit);
 
diff --git a/src/citra_qt/configure_graphics.cpp b/src/citra_qt/configure_graphics.cpp
index cea7db388..54f799b47 100644
--- a/src/citra_qt/configure_graphics.cpp
+++ b/src/citra_qt/configure_graphics.cpp
@@ -18,10 +18,81 @@ ConfigureGraphics::ConfigureGraphics(QWidget* parent)
 
 ConfigureGraphics::~ConfigureGraphics() {}
 
+enum class Resolution : int {
+    Auto,
+    Scale1x,
+    Scale2x,
+    Scale3x,
+    Scale4x,
+    Scale5x,
+    Scale6x,
+    Scale7x,
+    Scale8x,
+    Scale9x,
+    Scale10x,
+};
+
+float ToResolutionFactor(Resolution option) {
+    switch (option) {
+    case Resolution::Auto:
+        return 0.f;
+    case Resolution::Scale1x:
+        return 1.f;
+    case Resolution::Scale2x:
+        return 2.f;
+    case Resolution::Scale3x:
+        return 3.f;
+    case Resolution::Scale4x:
+        return 4.f;
+    case Resolution::Scale5x:
+        return 5.f;
+    case Resolution::Scale6x:
+        return 6.f;
+    case Resolution::Scale7x:
+        return 7.f;
+    case Resolution::Scale8x:
+        return 8.f;
+    case Resolution::Scale9x:
+        return 9.f;
+    case Resolution::Scale10x:
+        return 10.f;
+    }
+    return 0.f;
+}
+
+Resolution FromResolutionFactor(float factor) {
+    if (factor == 0.f) {
+        return Resolution::Auto;
+    } else if (factor == 1.f) {
+        return Resolution::Scale1x;
+    } else if (factor == 2.f) {
+        return Resolution::Scale2x;
+    } else if (factor == 3.f) {
+        return Resolution::Scale3x;
+    } else if (factor == 4.f) {
+        return Resolution::Scale4x;
+    } else if (factor == 5.f) {
+        return Resolution::Scale5x;
+    } else if (factor == 6.f) {
+        return Resolution::Scale6x;
+    } else if (factor == 7.f) {
+        return Resolution::Scale7x;
+    } else if (factor == 8.f) {
+        return Resolution::Scale8x;
+    } else if (factor == 9.f) {
+        return Resolution::Scale9x;
+    } else if (factor == 10.f) {
+        return Resolution::Scale10x;
+    }
+    return Resolution::Auto;
+}
+
 void ConfigureGraphics::setConfiguration() {
     ui->toggle_hw_renderer->setChecked(Settings::values.use_hw_renderer);
+    ui->resolution_factor_combobox->setEnabled(Settings::values.use_hw_renderer);
     ui->toggle_shader_jit->setChecked(Settings::values.use_shader_jit);
-    ui->toggle_scaled_resolution->setChecked(Settings::values.use_scaled_resolution);
+    ui->resolution_factor_combobox->setCurrentIndex(
+        static_cast<int>(FromResolutionFactor(Settings::values.resolution_factor)));
     ui->toggle_vsync->setChecked(Settings::values.use_vsync);
     ui->toggle_framelimit->setChecked(Settings::values.toggle_framelimit);
     ui->layout_combobox->setCurrentIndex(static_cast<int>(Settings::values.layout_option));
@@ -31,7 +102,8 @@ void ConfigureGraphics::setConfiguration() {
 void ConfigureGraphics::applyConfiguration() {
     Settings::values.use_hw_renderer = ui->toggle_hw_renderer->isChecked();
     Settings::values.use_shader_jit = ui->toggle_shader_jit->isChecked();
-    Settings::values.use_scaled_resolution = ui->toggle_scaled_resolution->isChecked();
+    Settings::values.resolution_factor =
+        ToResolutionFactor(static_cast<Resolution>(ui->resolution_factor_combobox->currentIndex()));
     Settings::values.use_vsync = ui->toggle_vsync->isChecked();
     Settings::values.toggle_framelimit = ui->toggle_framelimit->isChecked();
     Settings::values.layout_option =
diff --git a/src/citra_qt/configure_graphics.ui b/src/citra_qt/configure_graphics.ui
index 964aa0bbd..62021fe22 100644
--- a/src/citra_qt/configure_graphics.ui
+++ b/src/citra_qt/configure_graphics.ui
@@ -37,13 +37,6 @@
          </widget>
         </item>
         <item>
-         <widget class="QCheckBox" name="toggle_scaled_resolution">
-          <property name="text">
-           <string>Enable scaled resolution</string>
-          </property>
-         </widget>
-        </item>
-        <item>
          <widget class="QCheckBox" name="toggle_vsync">
           <property name="text">
            <string>Enable V-Sync</string>
@@ -57,6 +50,76 @@
           </property>
          </widget>
         </item>
+        <item>
+          <layout class="QHBoxLayout" name="horizontalLayout">
+            <item>
+              <widget class="QLabel" name="label">
+                <property name="text">
+                  <string>Internal Resolution:</string>
+                </property>
+              </widget>
+            </item>
+            <item>
+              <widget class="QComboBox" name="resolution_factor_combobox">
+                <item>
+                  <property name="text">
+                    <string notr="true">Auto (Window Size)</string>
+                  </property>
+                </item>
+                <item>
+                  <property name="text">
+                    <string notr="true">Native (400x240)</string>
+                  </property>
+                </item>
+                <item>
+                  <property name="text">
+                    <string notr="true">2x Native (800x480)</string>
+                  </property>
+                </item>
+                <item>
+                  <property name="text">
+                    <string notr="true">3x Native (1200x720)</string>
+                  </property>
+                </item>
+                <item>
+                  <property name="text">
+                    <string notr="true">4x Native (1600x960)</string>
+                  </property>
+                </item>
+                <item>
+                  <property name="text">
+                    <string notr="true">5x Native (2000x1200)</string>
+                  </property>
+                </item>
+                <item>
+                  <property name="text">
+                    <string notr="true">6x Native (2400x1440)</string>
+                  </property>
+                </item>
+                <item>
+                  <property name="text">
+                    <string notr="true">7x Native (2800x1680)</string>
+                  </property>
+                </item>
+                <item>
+                  <property name="text">
+                    <string notr="true">8x Native (3200x1920)</string>
+                  </property>
+                </item>
+                <item>
+                  <property name="text">
+                    <string notr="true">9x Native (3600x2160)</string>
+                  </property>
+                </item>
+                <item>
+                  <property name="text">
+                    <string notr="true">10x Native (4000x2400)</string>
+                  </property>
+                </item>
+              </widget>
+            </item>
+          </layout>
+        </item>
        </layout>
       </widget>
      </item>
@@ -128,5 +191,12 @@
   </layout>
  </widget>
  <resources/>
- <connections/>
+ <connections>
+  <connection>
+   <sender>toggle_hw_renderer</sender>
+   <signal>toggled(bool)</signal>
+   <receiver>resolution_factor_combobox</receiver>
+   <slot>setEnabled(bool)</slot>
+  </connection>
+ </connections>
 </ui>
diff --git a/src/citra_qt/debugger/callstack.cpp b/src/citra_qt/debugger/callstack.cpp
index c1db93583..08d2e7a22 100644
--- a/src/citra_qt/debugger/callstack.cpp
+++ b/src/citra_qt/debugger/callstack.cpp
@@ -45,7 +45,6 @@ void CallstackWidget::OnDebugModeEntered() {
         if (ARM_Disasm::Decode(insn) == OP_BL) {
             std::string name;
             // ripped from disasm
-            u8 cond = (insn >> 28) & 0xf;
             u32 i_offset = insn & 0xffffff;
             // Sign-extend the 24-bit offset
             if ((i_offset >> 23) & 1)
diff --git a/src/citra_qt/debugger/graphics/graphics_cmdlists.cpp b/src/citra_qt/debugger/graphics/graphics_cmdlists.cpp
index dab529e3a..f5a2ec761 100644
--- a/src/citra_qt/debugger/graphics/graphics_cmdlists.cpp
+++ b/src/citra_qt/debugger/graphics/graphics_cmdlists.cpp
@@ -135,11 +135,6 @@ void GPUCommandListWidget::OnCommandDoubleClicked(const QModelIndex& index) {
             UNREACHABLE_MSG("Unknown texture command");
         }
 
-        const auto texture = Pica::g_state.regs.GetTextures()[texture_index];
-        const auto config = texture.config;
-        const auto format = texture.format;
-        const auto info = Pica::DebugUtils::TextureInfo::FromPicaRegister(config, format);
-
         // TODO: Open a surface debugger
     }
 }
diff --git a/src/citra_qt/debugger/graphics/graphics_vertex_shader.cpp b/src/citra_qt/debugger/graphics/graphics_vertex_shader.cpp
index b75b94ef8..ff2e7e363 100644
--- a/src/citra_qt/debugger/graphics/graphics_vertex_shader.cpp
+++ b/src/citra_qt/debugger/graphics/graphics_vertex_shader.cpp
@@ -276,9 +276,6 @@ QVariant GraphicsVertexShaderModel::data(const QModelIndex& index, int role) con
                         output << 'b' << instr.flow_control.bool_uniform_id << ' ';
                     }
 
-                    u32 target_addr = instr.flow_control.dest_offset;
-                    u32 target_addr_else = instr.flow_control.dest_offset;
-
                     if (opcode_info.subtype & OpCode::Info::HasAlternative) {
                         output << "else jump to 0x" << std::setw(4) << std::right
                                << std::setfill('0') << std::hex
@@ -473,7 +470,6 @@ GraphicsVertexShaderWidget::GraphicsVertexShaderWidget(
 }
 
 void GraphicsVertexShaderWidget::OnBreakPointHit(Pica::DebugContext::Event event, void* data) {
-    auto input = static_cast<Pica::Shader::InputVertex*>(data);
     if (event == Pica::DebugContext::Event::VertexShaderInvocation) {
         Reload(true, data);
     } else {
diff --git a/src/citra_qt/debugger/wait_tree.cpp b/src/citra_qt/debugger/wait_tree.cpp
index 1d2de5185..b6ecf3819 100644
--- a/src/citra_qt/debugger/wait_tree.cpp
+++ b/src/citra_qt/debugger/wait_tree.cpp
@@ -153,7 +153,8 @@ QString WaitTreeThread::GetText() const {
     case THREADSTATUS_WAIT_SLEEP:
         status = tr("sleeping");
         break;
-    case THREADSTATUS_WAIT_SYNCH:
+    case THREADSTATUS_WAIT_SYNCH_ALL:
+    case THREADSTATUS_WAIT_SYNCH_ANY:
         status = tr("waiting for objects");
         break;
     case THREADSTATUS_DORMANT:
@@ -180,7 +181,8 @@ QColor WaitTreeThread::GetColor() const {
         return QColor(Qt::GlobalColor::darkRed);
     case THREADSTATUS_WAIT_SLEEP:
         return QColor(Qt::GlobalColor::darkYellow);
-    case THREADSTATUS_WAIT_SYNCH:
+    case THREADSTATUS_WAIT_SYNCH_ALL:
+    case THREADSTATUS_WAIT_SYNCH_ANY:
         return QColor(Qt::GlobalColor::red);
     case THREADSTATUS_DORMANT:
         return QColor(Qt::GlobalColor::darkCyan);
@@ -228,7 +230,8 @@ std::vector<std::unique_ptr<WaitTreeItem>> WaitTreeThread::GetChildren() const {
     } else {
         list.push_back(std::make_unique<WaitTreeMutexList>(thread.held_mutexes));
     }
-    if (thread.status == THREADSTATUS_WAIT_SYNCH) {
+    if (thread.status == THREADSTATUS_WAIT_SYNCH_ANY ||
+        thread.status == THREADSTATUS_WAIT_SYNCH_ALL) {
         list.push_back(std::make_unique<WaitTreeObjectList>(thread.wait_objects,
                                                             thread.IsSleepingOnWaitAll()));
     }
diff --git a/src/citra_qt/util/spinbox.cpp b/src/citra_qt/util/spinbox.cpp
index feb0ea1b3..212709007 100644
--- a/src/citra_qt/util/spinbox.cpp
+++ b/src/citra_qt/util/spinbox.cpp
@@ -165,13 +165,6 @@ void CSpinBox::UpdateText() {
         // Uppercase digits greater than 9.
         mask += ">";
 
-        // The greatest signed 64-bit number has 19 decimal digits.
-        // TODO: Could probably make this more generic with some logarithms.
-        // For reference, unsigned 64-bit can have up to 20 decimal digits.
-        int digits = (num_digits != 0)
-                         ? num_digits
-                         : (base == 16) ? 16 : (base == 10) ? 19 : 0xFF; // fallback case...
-
         // Match num_digits digits
         // Digits irrelevant to the chosen number base are filtered in the validator
         mask += QString("H").repeated(std::max(num_digits, 1));
diff --git a/src/common/CMakeLists.txt b/src/common/CMakeLists.txt
index 5aecf6e6e..a7a4a688c 100644
--- a/src/common/CMakeLists.txt
+++ b/src/common/CMakeLists.txt
@@ -46,6 +46,7 @@ set(HEADERS
             microprofileui.h
             platform.h
             profiler_reporting.h
+            quaternion.h
             scm_rev.h
             scope_exit.h
             string_util.h
diff --git a/src/common/logging/backend.cpp b/src/common/logging/backend.cpp
index 3ea102229..2ef3e6b05 100644
--- a/src/common/logging/backend.cpp
+++ b/src/common/logging/backend.cpp
@@ -45,6 +45,7 @@ namespace Log {
     SUB(Service, LDR)                                                                              \
     SUB(Service, MIC)                                                                              \
     SUB(Service, NDM)                                                                              \
+    SUB(Service, NFC)                                                                              \
     SUB(Service, NIM)                                                                              \
     SUB(Service, NWM)                                                                              \
     SUB(Service, CAM)                                                                              \
diff --git a/src/common/logging/log.h b/src/common/logging/log.h
index 9d8c18d8e..4330ef879 100644
--- a/src/common/logging/log.h
+++ b/src/common/logging/log.h
@@ -62,6 +62,7 @@ enum class Class : ClassType {
     Service_LDR,       ///< The LDR (3ds dll loader) service
     Service_MIC,       ///< The MIC (Microphone) service
     Service_NDM,       ///< The NDM (Network daemon manager) service
+    Service_NFC,       ///< The NFC service
     Service_NIM,       ///< The NIM (Network interface manager) service
     Service_NWM,       ///< The NWM (Network wlan manager) service
     Service_CAM,       ///< The CAM (Camera) service
diff --git a/src/common/math_util.h b/src/common/math_util.h
index cdeaeb733..45a1ed367 100644
--- a/src/common/math_util.h
+++ b/src/common/math_util.h
@@ -10,6 +10,8 @@
 
 namespace MathUtil {
 
+static constexpr float PI = 3.14159265f;
+
 inline bool IntervalsIntersect(unsigned start0, unsigned length0, unsigned start1,
                                unsigned length1) {
     return (std::max(start0, start1) < std::min(start0 + length0, start1 + length1));
diff --git a/src/common/quaternion.h b/src/common/quaternion.h
new file mode 100644
index 000000000..84ac82ed3
--- /dev/null
+++ b/src/common/quaternion.h
@@ -0,0 +1,44 @@
+// Copyright 2016 Citra Emulator Project
+// Licensed under GPLv2 or any later version
+// Refer to the license.txt file included.
+
+#pragma once
+
+#include "common/vector_math.h"
+
+namespace Math {
+
+template <typename T>
+class Quaternion {
+public:
+    Math::Vec3<T> xyz;
+    T w;
+
+    Quaternion<decltype(-T{})> Inverse() const {
+        return {-xyz, w};
+    }
+
+    Quaternion<decltype(T{} + T{})> operator+(const Quaternion& other) const {
+        return {xyz + other.xyz, w + other.w};
+    }
+
+    Quaternion<decltype(T{} - T{})> operator-(const Quaternion& other) const {
+        return {xyz - other.xyz, w - other.w};
+    }
+
+    Quaternion<decltype(T{} * T{} - T{} * T{})> operator*(const Quaternion& other) const {
+        return {xyz * other.w + other.xyz * w + Cross(xyz, other.xyz),
+                w * other.w - Dot(xyz, other.xyz)};
+    }
+};
+
+template <typename T>
+auto QuaternionRotate(const Quaternion<T>& q, const Math::Vec3<T>& v) {
+    return v + 2 * Cross(q.xyz, Cross(q.xyz, v) + v * q.w);
+}
+
+inline Quaternion<float> MakeQuaternion(const Math::Vec3<float>& axis, float angle) {
+    return {axis * std::sin(angle / 2), std::cos(angle / 2)};
+}
+
+} // namspace Math
diff --git a/src/common/thread.h b/src/common/thread.h
index 9c08be7e3..fa475ab51 100644
--- a/src/common/thread.h
+++ b/src/common/thread.h
@@ -4,6 +4,7 @@
 
 #pragma once
 
+#include <chrono>
 #include <condition_variable>
 #include <cstddef>
 #include <mutex>
@@ -54,6 +55,15 @@ public:
         is_set = false;
     }
 
+    template <class Clock, class Duration>
+    bool WaitUntil(const std::chrono::time_point<Clock, Duration>& time) {
+        std::unique_lock<std::mutex> lk(mutex);
+        if (!condvar.wait_until(lk, time, [this] { return is_set; }))
+            return false;
+        is_set = false;
+        return true;
+    }
+
     void Reset() {
         std::unique_lock<std::mutex> lk(mutex);
         // no other action required, since wait loops on the predicate and any lingering signal will
diff --git a/src/common/vector_math.h b/src/common/vector_math.h
index a57d86d88..7ca8e15f5 100644
--- a/src/common/vector_math.h
+++ b/src/common/vector_math.h
@@ -186,6 +186,18 @@ Vec2<T> operator*(const V& f, const Vec2<T>& vec) {
 
 typedef Vec2<float> Vec2f;
 
+template <>
+inline float Vec2<float>::Length() const {
+    return std::sqrt(x * x + y * y);
+}
+
+template <>
+inline float Vec2<float>::Normalize() {
+    float length = Length();
+    *this /= length;
+    return length;
+}
+
 template <typename T>
 class Vec3 {
 public:
@@ -388,6 +400,13 @@ inline Vec3<float> Vec3<float>::Normalized() const {
     return *this / Length();
 }
 
+template <>
+inline float Vec3<float>::Normalize() {
+    float length = Length();
+    *this /= length;
+    return length;
+}
+
 typedef Vec3<float> Vec3f;
 
 template <typename T>
diff --git a/src/core/CMakeLists.txt b/src/core/CMakeLists.txt
index 3621449b3..4c5b633e0 100644
--- a/src/core/CMakeLists.txt
+++ b/src/core/CMakeLists.txt
@@ -31,6 +31,7 @@ set(SRCS
             file_sys/savedata_archive.cpp
             frontend/emu_window.cpp
             frontend/key_map.cpp
+            frontend/motion_emu.cpp
             gdbstub/gdbstub.cpp
             hle/config_mem.cpp
             hle/applets/applet.cpp
@@ -202,6 +203,7 @@ set(HEADERS
             file_sys/savedata_archive.h
             frontend/emu_window.h
             frontend/key_map.h
+            frontend/motion_emu.h
             gdbstub/gdbstub.h
             hle/config_mem.h
             hle/function_wrappers.h
diff --git a/src/core/frontend/emu_window.cpp b/src/core/frontend/emu_window.cpp
index f6f90f9e1..1541cc39d 100644
--- a/src/core/frontend/emu_window.cpp
+++ b/src/core/frontend/emu_window.cpp
@@ -5,6 +5,7 @@
 #include <algorithm>
 #include <cmath>
 #include "common/assert.h"
+#include "common/profiler_reporting.h"
 #include "core/frontend/emu_window.h"
 #include "core/frontend/key_map.h"
 #include "video_core/video_core.h"
@@ -89,6 +90,30 @@ void EmuWindow::TouchMoved(unsigned framebuffer_x, unsigned framebuffer_y) {
     TouchPressed(framebuffer_x, framebuffer_y);
 }
 
+void EmuWindow::AccelerometerChanged(float x, float y, float z) {
+    constexpr float coef = 512;
+
+    std::lock_guard<std::mutex> lock(accel_mutex);
+
+    // TODO(wwylele): do a time stretch as it in GyroscopeChanged
+    // The time stretch formula should be like
+    // stretched_vector = (raw_vector - gravity) * stretch_ratio + gravity
+    accel_x = x * coef;
+    accel_y = y * coef;
+    accel_z = z * coef;
+}
+
+void EmuWindow::GyroscopeChanged(float x, float y, float z) {
+    constexpr float FULL_FPS = 60;
+    float coef = GetGyroscopeRawToDpsCoefficient();
+    float stretch =
+        FULL_FPS / Common::Profiling::GetTimingResultsAggregator()->GetAggregatedResults().fps;
+    std::lock_guard<std::mutex> lock(gyro_mutex);
+    gyro_x = x * coef * stretch;
+    gyro_y = y * coef * stretch;
+    gyro_z = z * coef * stretch;
+}
+
 void EmuWindow::UpdateCurrentFramebufferLayout(unsigned width, unsigned height) {
     Layout::FramebufferLayout layout;
     switch (Settings::values.layout_option) {
diff --git a/src/core/frontend/emu_window.h b/src/core/frontend/emu_window.h
index 835c4d500..1ba64c92b 100644
--- a/src/core/frontend/emu_window.h
+++ b/src/core/frontend/emu_window.h
@@ -4,6 +4,7 @@
 
 #pragma once
 
+#include <mutex>
 #include <tuple>
 #include <utility>
 #include "common/common_types.h"
@@ -93,6 +94,27 @@ public:
     void TouchMoved(unsigned framebuffer_x, unsigned framebuffer_y);
 
     /**
+     * Signal accelerometer state has changed.
+     * @param x X-axis accelerometer value
+     * @param y Y-axis accelerometer value
+     * @param z Z-axis accelerometer value
+     * @note all values are in unit of g (gravitational acceleration).
+     *    e.g. x = 1.0 means 9.8m/s^2 in x direction.
+     * @see GetAccelerometerState for axis explanation.
+     */
+    void AccelerometerChanged(float x, float y, float z);
+
+    /**
+     * Signal gyroscope state has changed.
+     * @param x X-axis accelerometer value
+     * @param y Y-axis accelerometer value
+     * @param z Z-axis accelerometer value
+     * @note all values are in deg/sec.
+     * @see GetGyroscopeState for axis explanation.
+     */
+    void GyroscopeChanged(float x, float y, float z);
+
+    /**
      * Gets the current pad state (which buttons are pressed).
      * @note This should be called by the core emu thread to get a state set by the window thread.
      * @note This doesn't include analog input like circle pad direction
@@ -134,12 +156,11 @@ public:
      *   1 unit of return value = 1/512 g (measured by hw test),
      *   where g is the gravitational acceleration (9.8 m/sec2).
      * @note This should be called by the core emu thread to get a state set by the window thread.
-     * @todo Implement accelerometer input in front-end.
      * @return std::tuple of (x, y, z)
      */
-    std::tuple<s16, s16, s16> GetAccelerometerState() const {
-        // stubbed
-        return std::make_tuple(0, -512, 0);
+    std::tuple<s16, s16, s16> GetAccelerometerState() {
+        std::lock_guard<std::mutex> lock(accel_mutex);
+        return std::make_tuple(accel_x, accel_y, accel_z);
     }
 
     /**
@@ -153,12 +174,11 @@ public:
      *   1 unit of return value = (1/coef) deg/sec,
      *   where coef is the return value of GetGyroscopeRawToDpsCoefficient().
      * @note This should be called by the core emu thread to get a state set by the window thread.
-     * @todo Implement gyroscope input in front-end.
      * @return std::tuple of (x, y, z)
      */
-    std::tuple<s16, s16, s16> GetGyroscopeState() const {
-        // stubbed
-        return std::make_tuple(0, 0, 0);
+    std::tuple<s16, s16, s16> GetGyroscopeState() {
+        std::lock_guard<std::mutex> lock(gyro_mutex);
+        return std::make_tuple(gyro_x, gyro_y, gyro_z);
     }
 
     /**
@@ -216,6 +236,12 @@ protected:
         circle_pad_x = 0;
         circle_pad_y = 0;
         touch_pressed = false;
+        accel_x = 0;
+        accel_y = -512;
+        accel_z = 0;
+        gyro_x = 0;
+        gyro_y = 0;
+        gyro_z = 0;
     }
     virtual ~EmuWindow() {}
 
@@ -281,6 +307,16 @@ private:
     s16 circle_pad_x; ///< Circle pad X-position in native 3DS pixel coordinates (-156 - 156)
     s16 circle_pad_y; ///< Circle pad Y-position in native 3DS pixel coordinates (-156 - 156)
 
+    std::mutex accel_mutex;
+    s16 accel_x; ///< Accelerometer X-axis value in native 3DS units
+    s16 accel_y; ///< Accelerometer Y-axis value in native 3DS units
+    s16 accel_z; ///< Accelerometer Z-axis value in native 3DS units
+
+    std::mutex gyro_mutex;
+    s16 gyro_x; ///< Gyroscope X-axis value in native 3DS units
+    s16 gyro_y; ///< Gyroscope Y-axis value in native 3DS units
+    s16 gyro_z; ///< Gyroscope Z-axis value in native 3DS units
+
     /**
      * Clip the provided coordinates to be inside the touchscreen area.
      */
diff --git a/src/core/frontend/motion_emu.cpp b/src/core/frontend/motion_emu.cpp
new file mode 100644
index 000000000..9a5b3185d
--- /dev/null
+++ b/src/core/frontend/motion_emu.cpp
@@ -0,0 +1,89 @@
+// Copyright 2016 Citra Emulator Project
+// Licensed under GPLv2 or any later version
+// Refer to the license.txt file included.
+
+#include "common/math_util.h"
+#include "common/quaternion.h"
+#include "core/frontend/emu_window.h"
+#include "core/frontend/motion_emu.h"
+
+namespace Motion {
+
+static constexpr int update_millisecond = 100;
+static constexpr auto update_duration =
+    std::chrono::duration_cast<std::chrono::steady_clock::duration>(
+        std::chrono::milliseconds(update_millisecond));
+
+MotionEmu::MotionEmu(EmuWindow& emu_window)
+    : motion_emu_thread(&MotionEmu::MotionEmuThread, this, std::ref(emu_window)) {}
+
+MotionEmu::~MotionEmu() {
+    if (motion_emu_thread.joinable()) {
+        shutdown_event.Set();
+        motion_emu_thread.join();
+    }
+}
+
+void MotionEmu::MotionEmuThread(EmuWindow& emu_window) {
+    auto update_time = std::chrono::steady_clock::now();
+    Math::Quaternion<float> q = MakeQuaternion(Math::Vec3<float>(), 0);
+    Math::Quaternion<float> old_q;
+
+    while (!shutdown_event.WaitUntil(update_time)) {
+        update_time += update_duration;
+        old_q = q;
+
+        {
+            std::lock_guard<std::mutex> guard(tilt_mutex);
+
+            // Find the quaternion describing current 3DS tilting
+            q = MakeQuaternion(Math::MakeVec(-tilt_direction.y, 0.0f, tilt_direction.x),
+                               tilt_angle);
+        }
+
+        auto inv_q = q.Inverse();
+
+        // Set the gravity vector in world space
+        auto gravity = Math::MakeVec(0.0f, -1.0f, 0.0f);
+
+        // Find the angular rate vector in world space
+        auto angular_rate = ((q - old_q) * inv_q).xyz * 2;
+        angular_rate *= 1000 / update_millisecond / MathUtil::PI * 180;
+
+        // Transform the two vectors from world space to 3DS space
+        gravity = QuaternionRotate(inv_q, gravity);
+        angular_rate = QuaternionRotate(inv_q, angular_rate);
+
+        // Update the sensor state
+        emu_window.AccelerometerChanged(gravity.x, gravity.y, gravity.z);
+        emu_window.GyroscopeChanged(angular_rate.x, angular_rate.y, angular_rate.z);
+    }
+}
+
+void MotionEmu::BeginTilt(int x, int y) {
+    mouse_origin = Math::MakeVec(x, y);
+    is_tilting = true;
+}
+
+void MotionEmu::Tilt(int x, int y) {
+    constexpr float SENSITIVITY = 0.01f;
+    auto mouse_move = Math::MakeVec(x, y) - mouse_origin;
+    if (is_tilting) {
+        std::lock_guard<std::mutex> guard(tilt_mutex);
+        if (mouse_move.x == 0 && mouse_move.y == 0) {
+            tilt_angle = 0;
+        } else {
+            tilt_direction = mouse_move.Cast<float>();
+            tilt_angle = MathUtil::Clamp(tilt_direction.Normalize() * SENSITIVITY, 0.0f,
+                                         MathUtil::PI * 0.5f);
+        }
+    }
+}
+
+void MotionEmu::EndTilt() {
+    std::lock_guard<std::mutex> guard(tilt_mutex);
+    tilt_angle = 0;
+    is_tilting = false;
+}
+
+} // namespace Motion
diff --git a/src/core/frontend/motion_emu.h b/src/core/frontend/motion_emu.h
new file mode 100644
index 000000000..99d41a726
--- /dev/null
+++ b/src/core/frontend/motion_emu.h
@@ -0,0 +1,52 @@
+// Copyright 2016 Citra Emulator Project
+// Licensed under GPLv2 or any later version
+// Refer to the license.txt file included.
+
+#pragma once
+#include "common/thread.h"
+#include "common/vector_math.h"
+
+class EmuWindow;
+
+namespace Motion {
+
+class MotionEmu final {
+public:
+    MotionEmu(EmuWindow& emu_window);
+    ~MotionEmu();
+
+    /**
+     * Signals that a motion sensor tilt has begun.
+     * @param x the x-coordinate of the cursor
+     * @param y the y-coordinate of the cursor
+     */
+    void BeginTilt(int x, int y);
+
+    /**
+     * Signals that a motion sensor tilt is occurring.
+     * @param x the x-coordinate of the cursor
+     * @param y the y-coordinate of the cursor
+     */
+    void Tilt(int x, int y);
+
+    /**
+     * Signals that a motion sensor tilt has ended.
+     */
+    void EndTilt();
+
+private:
+    Math::Vec2<int> mouse_origin;
+
+    std::mutex tilt_mutex;
+    Math::Vec2<float> tilt_direction;
+    float tilt_angle = 0;
+
+    bool is_tilting = false;
+
+    Common::Event shutdown_event;
+    std::thread motion_emu_thread;
+
+    void MotionEmuThread(EmuWindow& emu_window);
+};
+
+} // namespace Motion
diff --git a/src/core/gdbstub/gdbstub.cpp b/src/core/gdbstub/gdbstub.cpp
index d88e25073..5cf45ada5 100644
--- a/src/core/gdbstub/gdbstub.cpp
+++ b/src/core/gdbstub/gdbstub.cpp
@@ -57,7 +57,6 @@ const u32 SIGTERM = 15;
 const u32 MSG_WAITALL = 8;
 #endif
 
-const u32 R0_REGISTER = 0;
 const u32 R15_REGISTER = 15;
 const u32 CPSR_REGISTER = 25;
 const u32 FPSCR_REGISTER = 58;
@@ -816,10 +815,6 @@ static void RemoveBreakpoint() {
     auto addr_pos = std::find(start_offset, command_buffer + command_length, ',');
     PAddr addr = HexToInt(start_offset, static_cast<u32>(addr_pos - start_offset));
 
-    start_offset = addr_pos + 1;
-    u32 len =
-        HexToInt(start_offset, static_cast<u32>((command_buffer + command_length) - start_offset));
-
     if (type == BreakpointType::Access) {
         // Access is made up of Read and Write types, so add both breakpoints
         type = BreakpointType::Read;
diff --git a/src/core/hle/kernel/event.cpp b/src/core/hle/kernel/event.cpp
index a515f53f9..23f9df0d6 100644
--- a/src/core/hle/kernel/event.cpp
+++ b/src/core/hle/kernel/event.cpp
@@ -25,12 +25,12 @@ SharedPtr<Event> Event::Create(ResetType reset_type, std::string name) {
     return evt;
 }
 
-bool Event::ShouldWait() {
+bool Event::ShouldWait(Thread* thread) const {
     return !signaled;
 }
 
-void Event::Acquire() {
-    ASSERT_MSG(!ShouldWait(), "object unavailable!");
+void Event::Acquire(Thread* thread) {
+    ASSERT_MSG(!ShouldWait(thread), "object unavailable!");
 
     if (reset_type == ResetType::OneShot)
         signaled = false;
diff --git a/src/core/hle/kernel/event.h b/src/core/hle/kernel/event.h
index 28d955b1e..3e3673508 100644
--- a/src/core/hle/kernel/event.h
+++ b/src/core/hle/kernel/event.h
@@ -35,8 +35,8 @@ public:
     bool signaled;    ///< Whether the event has already been signaled
     std::string name; ///< Name of event (optional)
 
-    bool ShouldWait() override;
-    void Acquire() override;
+    bool ShouldWait(Thread* thread) const override;
+    void Acquire(Thread* thread) override;
 
     void WakeupAllWaitingThreads() override;
 
diff --git a/src/core/hle/kernel/kernel.cpp b/src/core/hle/kernel/kernel.cpp
index 1db8e102f..f599916f0 100644
--- a/src/core/hle/kernel/kernel.cpp
+++ b/src/core/hle/kernel/kernel.cpp
@@ -3,7 +3,6 @@
 // Refer to the license.txt file included.
 
 #include <algorithm>
-#include <boost/range/algorithm_ext/erase.hpp>
 #include "common/assert.h"
 #include "common/logging/log.h"
 #include "core/hle/config_mem.h"
@@ -28,32 +27,39 @@ void WaitObject::AddWaitingThread(SharedPtr<Thread> thread) {
 
 void WaitObject::RemoveWaitingThread(Thread* thread) {
     auto itr = std::find(waiting_threads.begin(), waiting_threads.end(), thread);
+    // If a thread passed multiple handles to the same object,
+    // the kernel might attempt to remove the thread from the object's
+    // waiting threads list multiple times.
     if (itr != waiting_threads.end())
         waiting_threads.erase(itr);
 }
 
 SharedPtr<Thread> WaitObject::GetHighestPriorityReadyThread() {
-    // Remove the threads that are ready or already running from our waitlist
-    boost::range::remove_erase_if(waiting_threads, [](const SharedPtr<Thread>& thread) {
-        return thread->status == THREADSTATUS_RUNNING || thread->status == THREADSTATUS_READY ||
-               thread->status == THREADSTATUS_DEAD;
-    });
-
-    // TODO(Subv): This call should be performed inside the loop below to check if an object can be
-    // acquired by a particular thread. This is useful for things like recursive locking of Mutexes.
-    if (ShouldWait())
-        return nullptr;
-
     Thread* candidate = nullptr;
     s32 candidate_priority = THREADPRIO_LOWEST + 1;
 
     for (const auto& thread : waiting_threads) {
+        // The list of waiting threads must not contain threads that are not waiting to be awakened.
+        ASSERT_MSG(thread->status == THREADSTATUS_WAIT_SYNCH_ANY ||
+                       thread->status == THREADSTATUS_WAIT_SYNCH_ALL,
+                   "Inconsistent thread statuses in waiting_threads");
+
         if (thread->current_priority >= candidate_priority)
             continue;
 
-        bool ready_to_run =
-            std::none_of(thread->wait_objects.begin(), thread->wait_objects.end(),
-                         [](const SharedPtr<WaitObject>& object) { return object->ShouldWait(); });
+        if (ShouldWait(thread.get()))
+            continue;
+
+        // A thread is ready to run if it's either in THREADSTATUS_WAIT_SYNCH_ANY or
+        // in THREADSTATUS_WAIT_SYNCH_ALL and the rest of the objects it is waiting on are ready.
+        bool ready_to_run = true;
+        if (thread->status == THREADSTATUS_WAIT_SYNCH_ALL) {
+            ready_to_run = std::none_of(thread->wait_objects.begin(), thread->wait_objects.end(),
+                                        [&thread](const SharedPtr<WaitObject>& object) {
+                                            return object->ShouldWait(thread.get());
+                                        });
+        }
+
         if (ready_to_run) {
             candidate = thread.get();
             candidate_priority = thread->current_priority;
@@ -66,7 +72,7 @@ SharedPtr<Thread> WaitObject::GetHighestPriorityReadyThread() {
 void WaitObject::WakeupAllWaitingThreads() {
     while (auto thread = GetHighestPriorityReadyThread()) {
         if (!thread->IsSleepingOnWaitAll()) {
-            Acquire();
+            Acquire(thread.get());
             // Set the output index of the WaitSynchronizationN call to the index of this object.
             if (thread->wait_set_output) {
                 thread->SetWaitSynchronizationOutput(thread->GetWaitObjectIndex(this));
@@ -74,18 +80,17 @@ void WaitObject::WakeupAllWaitingThreads() {
             }
         } else {
             for (auto& object : thread->wait_objects) {
-                object->Acquire();
-                object->RemoveWaitingThread(thread.get());
+                object->Acquire(thread.get());
             }
             // Note: This case doesn't update the output index of WaitSynchronizationN.
-            // Clear the thread's waitlist
-            thread->wait_objects.clear();
         }
 
+        for (auto& object : thread->wait_objects)
+            object->RemoveWaitingThread(thread.get());
+        thread->wait_objects.clear();
+
         thread->SetWaitSynchronizationResult(RESULT_SUCCESS);
         thread->ResumeFromWait();
-        // Note: Removing the thread from the object's waitlist will be
-        // done by GetHighestPriorityReadyThread.
     }
 }
 
diff --git a/src/core/hle/kernel/kernel.h b/src/core/hle/kernel/kernel.h
index 1e68f9cab..bb8b99bb5 100644
--- a/src/core/hle/kernel/kernel.h
+++ b/src/core/hle/kernel/kernel.h
@@ -132,25 +132,26 @@ using SharedPtr = boost::intrusive_ptr<T>;
 class WaitObject : public Object {
 public:
     /**
-     * Check if the current thread should wait until the object is available
+     * Check if the specified thread should wait until the object is available
+     * @param thread The thread about which we're deciding.
      * @return True if the current thread should wait due to this object being unavailable
      */
-    virtual bool ShouldWait() = 0;
+    virtual bool ShouldWait(Thread* thread) const = 0;
 
-    /// Acquire/lock the object if it is available
-    virtual void Acquire() = 0;
+    /// Acquire/lock the object for the specified thread if it is available
+    virtual void Acquire(Thread* thread) = 0;
 
     /**
      * Add a thread to wait on this object
      * @param thread Pointer to thread to add
      */
-    void AddWaitingThread(SharedPtr<Thread> thread);
+    virtual void AddWaitingThread(SharedPtr<Thread> thread);
 
     /**
      * Removes a thread from waiting on this object (e.g. if it was resumed already)
      * @param thread Pointer to thread to remove
      */
-    void RemoveWaitingThread(Thread* thread);
+    virtual void RemoveWaitingThread(Thread* thread);
 
     /**
      * Wake up all threads waiting on this object that can be awoken, in priority order,
diff --git a/src/core/hle/kernel/mutex.cpp b/src/core/hle/kernel/mutex.cpp
index 736944bae..cef961289 100644
--- a/src/core/hle/kernel/mutex.cpp
+++ b/src/core/hle/kernel/mutex.cpp
@@ -6,26 +6,18 @@
 #include <vector>
 #include <boost/range/algorithm_ext/erase.hpp>
 #include "common/assert.h"
+#include "core/core.h"
 #include "core/hle/kernel/kernel.h"
 #include "core/hle/kernel/mutex.h"
 #include "core/hle/kernel/thread.h"
 
 namespace Kernel {
 
-/**
- * Resumes a thread waiting for the specified mutex
- * @param mutex The mutex that some thread is waiting on
- */
-static void ResumeWaitingThread(Mutex* mutex) {
-    // Reset mutex lock thread handle, nothing is waiting
-    mutex->lock_count = 0;
-    mutex->holding_thread = nullptr;
-    mutex->WakeupAllWaitingThreads();
-}
-
 void ReleaseThreadMutexes(Thread* thread) {
     for (auto& mtx : thread->held_mutexes) {
-        ResumeWaitingThread(mtx.get());
+        mtx->lock_count = 0;
+        mtx->holding_thread = nullptr;
+        mtx->WakeupAllWaitingThreads();
     }
     thread->held_mutexes.clear();
 }
@@ -40,52 +32,74 @@ SharedPtr<Mutex> Mutex::Create(bool initial_locked, std::string name) {
     mutex->name = std::move(name);
     mutex->holding_thread = nullptr;
 
-    // Acquire mutex with current thread if initialized as locked...
+    // Acquire mutex with current thread if initialized as locked
     if (initial_locked)
-        mutex->Acquire();
+        mutex->Acquire(GetCurrentThread());
 
     return mutex;
 }
 
-bool Mutex::ShouldWait() {
-    auto thread = GetCurrentThread();
-    bool wait = lock_count > 0 && holding_thread != thread;
-
-    // If the holding thread of the mutex is lower priority than this thread, that thread should
-    // temporarily inherit this thread's priority
-    if (wait && thread->current_priority < holding_thread->current_priority)
-        holding_thread->BoostPriority(thread->current_priority);
-
-    return wait;
-}
-
-void Mutex::Acquire() {
-    Acquire(GetCurrentThread());
+bool Mutex::ShouldWait(Thread* thread) const {
+    return lock_count > 0 && thread != holding_thread;
 }
 
-void Mutex::Acquire(SharedPtr<Thread> thread) {
-    ASSERT_MSG(!ShouldWait(), "object unavailable!");
+void Mutex::Acquire(Thread* thread) {
+    ASSERT_MSG(!ShouldWait(thread), "object unavailable!");
 
-    // Actually "acquire" the mutex only if we don't already have it...
+    // Actually "acquire" the mutex only if we don't already have it
     if (lock_count == 0) {
+        priority = thread->current_priority;
         thread->held_mutexes.insert(this);
-        holding_thread = std::move(thread);
+        holding_thread = thread;
+        thread->UpdatePriority();
+        Core::System::GetInstance().PrepareReschedule();
     }
 
     lock_count++;
 }
 
 void Mutex::Release() {
-    // Only release if the mutex is held...
+    // Only release if the mutex is held
     if (lock_count > 0) {
         lock_count--;
 
-        // Yield to the next thread only if we've fully released the mutex...
+        // Yield to the next thread only if we've fully released the mutex
         if (lock_count == 0) {
             holding_thread->held_mutexes.erase(this);
-            ResumeWaitingThread(this);
+            holding_thread->UpdatePriority();
+            holding_thread = nullptr;
+            WakeupAllWaitingThreads();
+            Core::System::GetInstance().PrepareReschedule();
         }
     }
 }
 
+void Mutex::AddWaitingThread(SharedPtr<Thread> thread) {
+    WaitObject::AddWaitingThread(thread);
+    thread->pending_mutexes.insert(this);
+    UpdatePriority();
+}
+
+void Mutex::RemoveWaitingThread(Thread* thread) {
+    WaitObject::RemoveWaitingThread(thread);
+    thread->pending_mutexes.erase(this);
+    UpdatePriority();
+}
+
+void Mutex::UpdatePriority() {
+    if (!holding_thread)
+        return;
+
+    s32 best_priority = THREADPRIO_LOWEST;
+    for (auto& waiter : GetWaitingThreads()) {
+        if (waiter->current_priority < best_priority)
+            best_priority = waiter->current_priority;
+    }
+
+    if (best_priority != priority) {
+        priority = best_priority;
+        holding_thread->UpdatePriority();
+    }
+}
+
 } // namespace
diff --git a/src/core/hle/kernel/mutex.h b/src/core/hle/kernel/mutex.h
index 53c3dc1f1..c57adf400 100644
--- a/src/core/hle/kernel/mutex.h
+++ b/src/core/hle/kernel/mutex.h
@@ -35,17 +35,22 @@ public:
     }
 
     int lock_count;                   ///< Number of times the mutex has been acquired
+    u32 priority;                     ///< The priority of the mutex, used for priority inheritance.
     std::string name;                 ///< Name of mutex (optional)
     SharedPtr<Thread> holding_thread; ///< Thread that has acquired the mutex
 
-    bool ShouldWait() override;
-    void Acquire() override;
-
     /**
-     * Acquires the specified mutex for the specified thread
-     * @param thread Thread that will acquire the mutex
+     * Elevate the mutex priority to the best priority
+     * among the priorities of all its waiting threads.
      */
-    void Acquire(SharedPtr<Thread> thread);
+    void UpdatePriority();
+
+    bool ShouldWait(Thread* thread) const override;
+    void Acquire(Thread* thread) override;
+
+    void AddWaitingThread(SharedPtr<Thread> thread) override;
+    void RemoveWaitingThread(Thread* thread) override;
+
     void Release();
 
 private:
diff --git a/src/core/hle/kernel/semaphore.cpp b/src/core/hle/kernel/semaphore.cpp
index bf7600780..8bda2f75d 100644
--- a/src/core/hle/kernel/semaphore.cpp
+++ b/src/core/hle/kernel/semaphore.cpp
@@ -30,12 +30,13 @@ ResultVal<SharedPtr<Semaphore>> Semaphore::Create(s32 initial_count, s32 max_cou
     return MakeResult<SharedPtr<Semaphore>>(std::move(semaphore));
 }
 
-bool Semaphore::ShouldWait() {
+bool Semaphore::ShouldWait(Thread* thread) const {
     return available_count <= 0;
 }
 
-void Semaphore::Acquire() {
-    ASSERT_MSG(!ShouldWait(), "object unavailable!");
+void Semaphore::Acquire(Thread* thread) {
+    if (available_count <= 0)
+        return;
     --available_count;
 }
 
diff --git a/src/core/hle/kernel/semaphore.h b/src/core/hle/kernel/semaphore.h
index e01908a25..cde94f7cc 100644
--- a/src/core/hle/kernel/semaphore.h
+++ b/src/core/hle/kernel/semaphore.h
@@ -39,8 +39,8 @@ public:
     s32 available_count; ///< Number of free slots left in the semaphore
     std::string name;    ///< Name of semaphore (optional)
 
-    bool ShouldWait() override;
-    void Acquire() override;
+    bool ShouldWait(Thread* thread) const override;
+    void Acquire(Thread* thread) override;
 
     /**
      * Releases a certain number of slots from a semaphore.
diff --git a/src/core/hle/kernel/server_port.cpp b/src/core/hle/kernel/server_port.cpp
index 6c19aa7c0..fd3bbbcad 100644
--- a/src/core/hle/kernel/server_port.cpp
+++ b/src/core/hle/kernel/server_port.cpp
@@ -14,13 +14,13 @@ namespace Kernel {
 ServerPort::ServerPort() {}
 ServerPort::~ServerPort() {}
 
-bool ServerPort::ShouldWait() {
+bool ServerPort::ShouldWait(Thread* thread) const {
     // If there are no pending sessions, we wait until a new one is added.
     return pending_sessions.size() == 0;
 }
 
-void ServerPort::Acquire() {
-    ASSERT_MSG(!ShouldWait(), "object unavailable!");
+void ServerPort::Acquire(Thread* thread) {
+    ASSERT_MSG(!ShouldWait(thread), "object unavailable!");
 }
 
 std::tuple<SharedPtr<ServerPort>, SharedPtr<ClientPort>> ServerPort::CreatePortPair(
diff --git a/src/core/hle/kernel/server_port.h b/src/core/hle/kernel/server_port.h
index b0f8df62c..6f8bdb6a9 100644
--- a/src/core/hle/kernel/server_port.h
+++ b/src/core/hle/kernel/server_port.h
@@ -53,8 +53,8 @@ public:
     /// ServerSessions created from this port inherit a reference to this handler.
     std::shared_ptr<Service::SessionRequestHandler> hle_handler;
 
-    bool ShouldWait() override;
-    void Acquire() override;
+    bool ShouldWait(Thread* thread) const override;
+    void Acquire(Thread* thread) override;
 
 private:
     ServerPort();
diff --git a/src/core/hle/kernel/server_session.cpp b/src/core/hle/kernel/server_session.cpp
index 146458c1c..9447ff236 100644
--- a/src/core/hle/kernel/server_session.cpp
+++ b/src/core/hle/kernel/server_session.cpp
@@ -29,12 +29,12 @@ ResultVal<SharedPtr<ServerSession>> ServerSession::Create(
     return MakeResult<SharedPtr<ServerSession>>(std::move(server_session));
 }
 
-bool ServerSession::ShouldWait() {
+bool ServerSession::ShouldWait(Thread* thread) const {
     return !signaled;
 }
 
-void ServerSession::Acquire() {
-    ASSERT_MSG(!ShouldWait(), "object unavailable!");
+void ServerSession::Acquire(Thread* thread) {
+    ASSERT_MSG(!ShouldWait(thread), "object unavailable!");
     signaled = false;
 }
 
diff --git a/src/core/hle/kernel/server_session.h b/src/core/hle/kernel/server_session.h
index 458284a5d..c088b9a19 100644
--- a/src/core/hle/kernel/server_session.h
+++ b/src/core/hle/kernel/server_session.h
@@ -57,9 +57,9 @@ public:
      */
     ResultCode HandleSyncRequest();
 
-    bool ShouldWait() override;
+    bool ShouldWait(Thread* thread) const override;
 
-    void Acquire() override;
+    void Acquire(Thread* thread) override;
 
     std::string name; ///< The name of this session (optional)
     bool signaled;    ///< Whether there's new data available to this ServerSession
diff --git a/src/core/hle/kernel/thread.cpp b/src/core/hle/kernel/thread.cpp
index 5fb95dada..8c6fbcd04 100644
--- a/src/core/hle/kernel/thread.cpp
+++ b/src/core/hle/kernel/thread.cpp
@@ -27,12 +27,12 @@ namespace Kernel {
 /// Event type for the thread wake up event
 static int ThreadWakeupEventType;
 
-bool Thread::ShouldWait() {
+bool Thread::ShouldWait(Thread* thread) const {
     return status != THREADSTATUS_DEAD;
 }
 
-void Thread::Acquire() {
-    ASSERT_MSG(!ShouldWait(), "object unavailable!");
+void Thread::Acquire(Thread* thread) {
+    ASSERT_MSG(!ShouldWait(thread), "object unavailable!");
 }
 
 // TODO(yuriks): This can be removed if Thread objects are explicitly pooled in the future, allowing
@@ -66,20 +66,6 @@ Thread* GetCurrentThread() {
 }
 
 /**
- * Check if a thread is waiting on the specified wait object
- * @param thread The thread to test
- * @param wait_object The object to test against
- * @return True if the thread is waiting, false otherwise
- */
-static bool CheckWait_WaitObject(const Thread* thread, WaitObject* wait_object) {
-    if (thread->status != THREADSTATUS_WAIT_SYNCH)
-        return false;
-
-    auto itr = std::find(thread->wait_objects.begin(), thread->wait_objects.end(), wait_object);
-    return itr != thread->wait_objects.end();
-}
-
-/**
  * Check if the specified thread is waiting on the specified address to be arbitrated
  * @param thread The thread to test
  * @param wait_address The address to test against
@@ -90,9 +76,6 @@ static bool CheckWait_AddressArbiter(const Thread* thread, VAddr wait_address) {
 }
 
 void Thread::Stop() {
-    // Release all the mutexes that this thread holds
-    ReleaseThreadMutexes(this);
-
     // Cancel any outstanding wakeup events for this thread
     CoreTiming::UnscheduleEvent(ThreadWakeupEventType, callback_handle);
     wakeup_callback_handle_table.Close(callback_handle);
@@ -114,6 +97,9 @@ 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.
     u32 tls_page = (tls_address - Memory::TLS_AREA_VADDR) / Memory::PAGE_SIZE;
     u32 tls_slot =
@@ -155,28 +141,6 @@ void ArbitrateAllThreads(u32 address) {
     }
 }
 
-/// Boost low priority threads (temporarily) that have been starved
-static void PriorityBoostStarvedThreads() {
-    u64 current_ticks = CoreTiming::GetTicks();
-
-    for (auto& thread : thread_list) {
-        // TODO(bunnei): Threads that have been waiting to be scheduled for `boost_ticks` (or
-        // longer) will have their priority temporarily adjusted to 1 higher than the highest
-        // priority thread to prevent thread starvation. This general behavior has been verified
-        // on hardware. However, this is almost certainly not perfect, and the real CTR OS scheduler
-        // should probably be reversed to verify this.
-
-        const u64 boost_timeout = 2000000; // Boost threads that have been ready for > this long
-
-        u64 delta = current_ticks - thread->last_running_ticks;
-
-        if (thread->status == THREADSTATUS_READY && delta > boost_timeout) {
-            const s32 priority = std::max(ready_queue.get_first()->current_priority - 1, 0);
-            thread->BoostPriority(priority);
-        }
-    }
-}
-
 /**
  * Switches the CPU's active thread context to that of the specified thread
  * @param new_thread The thread to switch to
@@ -199,8 +163,8 @@ static void SwitchContext(Thread* new_thread) {
 
     // Load context of new thread
     if (new_thread) {
-        DEBUG_ASSERT_MSG(new_thread->status == THREADSTATUS_READY,
-                         "Thread must be ready to become running.");
+        ASSERT_MSG(new_thread->status == THREADSTATUS_READY,
+                   "Thread must be ready to become running.");
 
         // Cancel any outstanding wakeup events for this thread
         CoreTiming::UnscheduleEvent(ThreadWakeupEventType, new_thread->callback_handle);
@@ -210,9 +174,6 @@ static void SwitchContext(Thread* new_thread) {
         ready_queue.remove(new_thread->current_priority, new_thread);
         new_thread->status = THREADSTATUS_RUNNING;
 
-        // Restores thread to its nominal priority if it has been temporarily changed
-        new_thread->current_priority = new_thread->nominal_priority;
-
         Core::CPU().LoadContext(new_thread->context);
         Core::CPU().SetCP15Register(CP15_THREAD_URO, new_thread->GetTLSAddress());
     } else {
@@ -248,14 +209,6 @@ void WaitCurrentThread_Sleep() {
     thread->status = THREADSTATUS_WAIT_SLEEP;
 }
 
-void WaitCurrentThread_WaitSynchronization(std::vector<SharedPtr<WaitObject>> wait_objects,
-                                           bool wait_set_output) {
-    Thread* thread = GetCurrentThread();
-    thread->wait_set_output = wait_set_output;
-    thread->wait_objects = std::move(wait_objects);
-    thread->status = THREADSTATUS_WAIT_SYNCH;
-}
-
 void WaitCurrentThread_ArbitrateAddress(VAddr wait_address) {
     Thread* thread = GetCurrentThread();
     thread->wait_address = wait_address;
@@ -281,7 +234,8 @@ static void ThreadWakeupCallback(u64 thread_handle, int cycles_late) {
         return;
     }
 
-    if (thread->status == THREADSTATUS_WAIT_SYNCH || thread->status == THREADSTATUS_WAIT_ARB) {
+    if (thread->status == THREADSTATUS_WAIT_SYNCH_ANY ||
+        thread->status == THREADSTATUS_WAIT_SYNCH_ALL || thread->status == THREADSTATUS_WAIT_ARB) {
         thread->wait_set_output = false;
         // Remove the thread from each of its waiting objects' waitlists
         for (auto& object : thread->wait_objects)
@@ -305,8 +259,11 @@ void Thread::WakeAfterDelay(s64 nanoseconds) {
 }
 
 void Thread::ResumeFromWait() {
+    ASSERT_MSG(wait_objects.empty(), "Thread is waking up while waiting for objects");
+
     switch (status) {
-    case THREADSTATUS_WAIT_SYNCH:
+    case THREADSTATUS_WAIT_SYNCH_ALL:
+    case THREADSTATUS_WAIT_SYNCH_ANY:
     case THREADSTATUS_WAIT_ARB:
     case THREADSTATUS_WAIT_SLEEP:
         break;
@@ -515,8 +472,21 @@ void Thread::SetPriority(s32 priority) {
     nominal_priority = current_priority = priority;
 }
 
+void Thread::UpdatePriority() {
+    s32 best_priority = nominal_priority;
+    for (auto& mutex : held_mutexes) {
+        if (mutex->priority < best_priority)
+            best_priority = mutex->priority;
+    }
+    BoostPriority(best_priority);
+}
+
 void Thread::BoostPriority(s32 priority) {
-    ready_queue.move(this, current_priority, priority);
+    // If thread was ready, adjust queues
+    if (status == THREADSTATUS_READY)
+        ready_queue.move(this, current_priority, priority);
+    else
+        ready_queue.prepare(priority);
     current_priority = priority;
 }
 
@@ -538,9 +508,11 @@ SharedPtr<Thread> SetupMainThread(u32 entry_point, s32 priority) {
     return thread;
 }
 
-void Reschedule() {
-    PriorityBoostStarvedThreads();
+bool HaveReadyThreads() {
+    return ready_queue.get_first() != nullptr;
+}
 
+void Reschedule() {
     Thread* cur = GetCurrentThread();
     Thread* next = PopNextReadyThread();
 
@@ -563,6 +535,12 @@ void Thread::SetWaitSynchronizationOutput(s32 output) {
     context.cpu_registers[1] = output;
 }
 
+s32 Thread::GetWaitObjectIndex(WaitObject* object) const {
+    ASSERT_MSG(!wait_objects.empty(), "Thread is not waiting for anything");
+    auto match = std::find(wait_objects.rbegin(), wait_objects.rend(), object);
+    return std::distance(match, wait_objects.rend()) - 1;
+}
+
 ////////////////////////////////////////////////////////////////////////////////////////////////////
 
 void ThreadingInit() {
diff --git a/src/core/hle/kernel/thread.h b/src/core/hle/kernel/thread.h
index c77ac644d..c557a2279 100644
--- a/src/core/hle/kernel/thread.h
+++ b/src/core/hle/kernel/thread.h
@@ -31,13 +31,14 @@ enum ThreadProcessorId : s32 {
 };
 
 enum ThreadStatus {
-    THREADSTATUS_RUNNING,    ///< Currently running
-    THREADSTATUS_READY,      ///< Ready to run
-    THREADSTATUS_WAIT_ARB,   ///< Waiting on an address arbiter
-    THREADSTATUS_WAIT_SLEEP, ///< Waiting due to a SleepThread SVC
-    THREADSTATUS_WAIT_SYNCH, ///< Waiting due to a WaitSynchronization SVC
-    THREADSTATUS_DORMANT,    ///< Created but not yet made ready
-    THREADSTATUS_DEAD        ///< Run to completion, or forcefully terminated
+    THREADSTATUS_RUNNING,        ///< Currently running
+    THREADSTATUS_READY,          ///< Ready to run
+    THREADSTATUS_WAIT_ARB,       ///< Waiting on an address arbiter
+    THREADSTATUS_WAIT_SLEEP,     ///< Waiting due to a SleepThread SVC
+    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_DORMANT,        ///< Created but not yet made ready
+    THREADSTATUS_DEAD            ///< Run to completion, or forcefully terminated
 };
 
 namespace Kernel {
@@ -72,8 +73,8 @@ public:
         return HANDLE_TYPE;
     }
 
-    bool ShouldWait() override;
-    void Acquire() override;
+    bool ShouldWait(Thread* thread) const override;
+    void Acquire(Thread* thread) override;
 
     /**
      * Gets the thread's current priority
@@ -90,6 +91,12 @@ public:
     void SetPriority(s32 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
      */
@@ -128,13 +135,14 @@ public:
 
     /**
      * Retrieves the index that this particular object occupies in the list of objects
-     * that the thread passed to WaitSynchronizationN.
+     * that the thread passed to WaitSynchronizationN, starting the search from the last element.
      * It is used to set the output value of WaitSynchronizationN when the thread is awakened.
+     * When a thread wakes up due to an object signal, the kernel will use the index of the last
+     * matching object in the wait objects list in case of having multiple instances of the same
+     * object in the list.
      * @param object Object to query the index of.
      */
-    s32 GetWaitObjectIndex(const WaitObject* object) const {
-        return wait_objects_index.at(object->GetObjectId());
-    }
+    s32 GetWaitObjectIndex(WaitObject* object) const;
 
     /**
      * Stops a thread, invalidating it from further use
@@ -152,10 +160,10 @@ public:
     /**
      * Returns whether this thread is waiting for all the objects in
      * its wait list to become ready, as a result of a WaitSynchronizationN call
-     * with wait_all = true, or a ReplyAndReceive call.
+     * with wait_all = true.
      */
     bool IsSleepingOnWaitAll() const {
-        return !wait_objects.empty();
+        return status == THREADSTATUS_WAIT_SYNCH_ALL;
     }
 
     ARM_Interface::ThreadContext context;
@@ -178,15 +186,15 @@ public:
     /// 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.
-    /// This is only populated when the thread should wait for all the objects to become ready.
+    /// Objects that the thread is waiting on, in the same order as they were
+    // passed to WaitSynchronization1/N.
     std::vector<SharedPtr<WaitObject>> wait_objects;
 
-    /// Mapping of Object ids to their position in the last waitlist that this object waited on.
-    boost::container::flat_map<int, s32> wait_objects_index;
-
     VAddr wait_address; ///< If waiting on an AddressArbiter, this is the arbitration address
 
     /// True if the WaitSynchronizationN output parameter should be set on thread wakeup.
@@ -211,6 +219,11 @@ private:
 SharedPtr<Thread> SetupMainThread(u32 entry_point, s32 priority);
 
 /**
+ * Returns whether there are any threads that are ready to run.
+ */
+bool HaveReadyThreads();
+
+/**
  * Reschedules to the next available thread (call after current thread is suspended)
  */
 void Reschedule();
@@ -238,15 +251,6 @@ Thread* GetCurrentThread();
 void WaitCurrentThread_Sleep();
 
 /**
- * Waits the current thread from a WaitSynchronization call
- * @param wait_objects Kernel objects that we are waiting on
- * @param wait_set_output If true, set the output parameter on thread wakeup (for
- * WaitSynchronizationN only)
- */
-void WaitCurrentThread_WaitSynchronization(std::vector<SharedPtr<WaitObject>> wait_objects,
-                                           bool wait_set_output);
-
-/**
  * Waits the current thread from an ArbitrateAddress call
  * @param wait_address Arbitration address used to resume from wait
  */
diff --git a/src/core/hle/kernel/timer.cpp b/src/core/hle/kernel/timer.cpp
index ec85b7ec8..60537f355 100644
--- a/src/core/hle/kernel/timer.cpp
+++ b/src/core/hle/kernel/timer.cpp
@@ -34,12 +34,12 @@ SharedPtr<Timer> Timer::Create(ResetType reset_type, std::string name) {
     return timer;
 }
 
-bool Timer::ShouldWait() {
+bool Timer::ShouldWait(Thread* thread) const {
     return !signaled;
 }
 
-void Timer::Acquire() {
-    ASSERT_MSG(!ShouldWait(), "object unavailable!");
+void Timer::Acquire(Thread* thread) {
+    ASSERT_MSG(!ShouldWait(thread), "object unavailable!");
 
     if (reset_type == ResetType::OneShot)
         signaled = false;
diff --git a/src/core/hle/kernel/timer.h b/src/core/hle/kernel/timer.h
index 2092165d4..c174f5664 100644
--- a/src/core/hle/kernel/timer.h
+++ b/src/core/hle/kernel/timer.h
@@ -39,8 +39,8 @@ public:
     u64 initial_delay;  ///< The delay until the timer fires for the first time
     u64 interval_delay; ///< The delay until the timer fires after the first time
 
-    bool ShouldWait() override;
-    void Acquire() override;
+    bool ShouldWait(Thread* thread) const override;
+    void Acquire(Thread* thread) override;
 
     void WakeupAllWaitingThreads() override;
 
diff --git a/src/core/hle/service/boss/boss.cpp b/src/core/hle/service/boss/boss.cpp
index 6ab16ccd5..e0de037f8 100644
--- a/src/core/hle/service/boss/boss.cpp
+++ b/src/core/hle/service/boss/boss.cpp
@@ -2,6 +2,7 @@
 // Licensed under GPLv2 or any later version
 // Refer to the license.txt file included.
 
+#include <cinttypes>
 #include "core/hle/service/boss/boss.h"
 #include "core/hle/service/boss/boss_p.h"
 #include "core/hle/service/boss/boss_u.h"
@@ -33,7 +34,8 @@ void InitializeSession(Service::Interface* self) {
     cmd_buff[0] = IPC::MakeHeader(0x1, 0x1, 0);
     cmd_buff[1] = RESULT_SUCCESS.raw;
 
-    LOG_WARNING(Service_BOSS, "(STUBBED) unk_param=0x%016X, translation=0x%08X, unk_param4=0x%08X",
+    LOG_WARNING(Service_BOSS,
+                "(STUBBED) unk_param=0x%016" PRIX64 ", translation=0x%08X, unk_param4=0x%08X",
                 unk_param, translation, unk_param4);
 }
 
diff --git a/src/core/hle/service/cfg/cfg.cpp b/src/core/hle/service/cfg/cfg.cpp
index 0bf59eb76..59dd6d1cd 100644
--- a/src/core/hle/service/cfg/cfg.cpp
+++ b/src/core/hle/service/cfg/cfg.cpp
@@ -84,7 +84,6 @@ struct ConsoleCountryInfo {
 static_assert(sizeof(ConsoleCountryInfo) == 4, "ConsoleCountryInfo must be exactly 4 bytes");
 }
 
-static const u64 CFG_SAVE_ID = 0x00010017;
 static const u64 CONSOLE_UNIQUE_ID = 0xDEADC0DE;
 static const ConsoleModelInfo CONSOLE_MODEL = {NINTENDO_3DS_XL, {0, 0, 0}};
 static const u8 CONSOLE_LANGUAGE = LANGUAGE_EN;
diff --git a/src/core/hle/service/mic_u.cpp b/src/core/hle/service/mic_u.cpp
index 4f1dd2fce..c62f8afc6 100644
--- a/src/core/hle/service/mic_u.cpp
+++ b/src/core/hle/service/mic_u.cpp
@@ -99,7 +99,8 @@ static void StartSampling(Interface* self) {
     is_sampling = true;
     LOG_WARNING(Service_MIC, "(STUBBED) called, encoding=%u, sample_rate=%u, "
                              "audio_buffer_offset=%d, audio_buffer_size=%u, audio_buffer_loop=%u",
-                encoding, sample_rate, audio_buffer_offset, audio_buffer_size, audio_buffer_loop);
+                static_cast<u32>(encoding), static_cast<u32>(sample_rate), audio_buffer_offset,
+                audio_buffer_size, audio_buffer_loop);
 }
 
 /**
@@ -114,7 +115,7 @@ static void AdjustSampling(Interface* self) {
     u32* cmd_buff = Kernel::GetCommandBuffer();
     sample_rate = static_cast<SampleRate>(cmd_buff[1] & 0xFF);
     cmd_buff[1] = RESULT_SUCCESS.raw; // No error
-    LOG_WARNING(Service_MIC, "(STUBBED) called, sample_rate=%u", sample_rate);
+    LOG_WARNING(Service_MIC, "(STUBBED) called, sample_rate=%u", static_cast<u32>(sample_rate));
 }
 
 /**
diff --git a/src/core/hle/service/nfc/nfc.cpp b/src/core/hle/service/nfc/nfc.cpp
index d9738c6a1..e248285f9 100644
--- a/src/core/hle/service/nfc/nfc.cpp
+++ b/src/core/hle/service/nfc/nfc.cpp
@@ -2,6 +2,7 @@
 // Licensed under GPLv2 or any later version
 // Refer to the license.txt file included.
 
+#include "core/hle/kernel/event.h"
 #include "core/hle/service/nfc/nfc.h"
 #include "core/hle/service/nfc/nfc_m.h"
 #include "core/hle/service/nfc/nfc_u.h"
@@ -9,9 +10,28 @@
 namespace Service {
 namespace NFC {
 
+static Kernel::SharedPtr<Kernel::Event> tag_in_range_event;
+
+void GetTagInRangeEvent(Interface* self) {
+    u32* cmd_buff = Kernel::GetCommandBuffer();
+
+    cmd_buff[0] = IPC::MakeHeader(0xB, 1, 2);
+    cmd_buff[1] = RESULT_SUCCESS.raw;
+    cmd_buff[2] = IPC::CopyHandleDesc();
+    cmd_buff[3] = Kernel::g_handle_table.Create(tag_in_range_event).MoveFrom();
+    LOG_WARNING(Service_NFC, "(STUBBED) called");
+}
+
 void Init() {
     AddService(new NFC_M());
     AddService(new NFC_U());
+
+    tag_in_range_event =
+        Kernel::Event::Create(Kernel::ResetType::OneShot, "NFC::tag_in_range_event");
+}
+
+void Shutdown() {
+    tag_in_range_event = nullptr;
 }
 
 } // namespace NFC
diff --git a/src/core/hle/service/nfc/nfc.h b/src/core/hle/service/nfc/nfc.h
index cd65a5fdc..b02354201 100644
--- a/src/core/hle/service/nfc/nfc.h
+++ b/src/core/hle/service/nfc/nfc.h
@@ -5,10 +5,27 @@
 #pragma once
 
 namespace Service {
+
+class Interface;
+
 namespace NFC {
 
+/**
+ * NFC::GetTagInRangeEvent service function
+ *  Inputs:
+ *      0 : Header code [0x000B0000]
+ *  Outputs:
+ *      1 : Result of function, 0 on success, otherwise error code
+ *      2 : Copy handle descriptor
+ *      3 : Event Handle
+ */
+void GetTagInRangeEvent(Interface* self);
+
 /// Initialize all NFC services.
 void Init();
 
+/// Shutdown all NFC services.
+void Shutdown();
+
 } // namespace NFC
 } // namespace Service
diff --git a/src/core/hle/service/nfc/nfc_m.cpp b/src/core/hle/service/nfc/nfc_m.cpp
index 717335c11..f43b4029a 100644
--- a/src/core/hle/service/nfc/nfc_m.cpp
+++ b/src/core/hle/service/nfc/nfc_m.cpp
@@ -2,6 +2,7 @@
 // Licensed under GPLv2 or any later version
 // Refer to the license.txt file included.
 
+#include "core/hle/service/nfc/nfc.h"
 #include "core/hle/service/nfc/nfc_m.h"
 
 namespace Service {
@@ -19,6 +20,7 @@ const Interface::FunctionInfo FunctionTable[] = {
     {0x00070000, nullptr, "LoadAmiiboData"},
     {0x00080000, nullptr, "ResetTagScanState"},
     {0x00090002, nullptr, "UpdateStoredAmiiboData"},
+    {0x000B0000, GetTagInRangeEvent, "GetTagInRangeEvent"},
     {0x000D0000, nullptr, "GetTagState"},
     {0x000F0000, nullptr, "CommunicationGetStatus"},
     {0x00100000, nullptr, "GetTagInfo2"},
diff --git a/src/core/hle/service/nfc/nfc_u.cpp b/src/core/hle/service/nfc/nfc_u.cpp
index deffb0b4f..4b5200ae8 100644
--- a/src/core/hle/service/nfc/nfc_u.cpp
+++ b/src/core/hle/service/nfc/nfc_u.cpp
@@ -2,6 +2,7 @@
 // Licensed under GPLv2 or any later version
 // Refer to the license.txt file included.
 
+#include "core/hle/service/nfc/nfc.h"
 #include "core/hle/service/nfc/nfc_u.h"
 
 namespace Service {
@@ -18,6 +19,7 @@ const Interface::FunctionInfo FunctionTable[] = {
     {0x00070000, nullptr, "LoadAmiiboData"},
     {0x00080000, nullptr, "ResetTagScanState"},
     {0x00090002, nullptr, "UpdateStoredAmiiboData"},
+    {0x000B0000, GetTagInRangeEvent, "GetTagInRangeEvent"},
     {0x000D0000, nullptr, "GetTagState"},
     {0x000F0000, nullptr, "CommunicationGetStatus"},
     {0x00100000, nullptr, "GetTagInfo2"},
diff --git a/src/core/hle/service/service.cpp b/src/core/hle/service/service.cpp
index 7e52a05d9..f3190e0fa 100644
--- a/src/core/hle/service/service.cpp
+++ b/src/core/hle/service/service.cpp
@@ -178,6 +178,7 @@ void Init() {
 /// Shutdown ServiceManager
 void Shutdown() {
     PTM::Shutdown();
+    NFC::Shutdown();
     NIM::Shutdown();
     NEWS::Shutdown();
     NDM::Shutdown();
diff --git a/src/core/hle/service/soc_u.cpp b/src/core/hle/service/soc_u.cpp
index c3918cdd0..dcc5c3c90 100644
--- a/src/core/hle/service/soc_u.cpp
+++ b/src/core/hle/service/soc_u.cpp
@@ -603,7 +603,6 @@ static void RecvFrom(Interface* self) {
     u32 socket_handle = cmd_buffer[1];
     u32 len = cmd_buffer[2];
     u32 flags = cmd_buffer[3];
-    socklen_t addr_len = static_cast<socklen_t>(cmd_buffer[4]);
 
     struct {
         u32 output_buffer_descriptor;
@@ -693,7 +692,6 @@ static void Poll(Interface* self) {
 static void GetSockName(Interface* self) {
     u32* cmd_buffer = Kernel::GetCommandBuffer();
     u32 socket_handle = cmd_buffer[1];
-    socklen_t ctr_len = cmd_buffer[2];
 
     // Memory address of the ctr_dest_addr structure
     VAddr ctr_dest_addr_addr = cmd_buffer[0x104 >> 2];
@@ -734,7 +732,6 @@ static void Shutdown(Interface* self) {
 static void GetPeerName(Interface* self) {
     u32* cmd_buffer = Kernel::GetCommandBuffer();
     u32 socket_handle = cmd_buffer[1];
-    socklen_t len = cmd_buffer[2];
 
     // Memory address of the ctr_dest_addr structure
     VAddr ctr_dest_addr_addr = cmd_buffer[0x104 >> 2];
@@ -765,7 +762,6 @@ static void Connect(Interface* self) {
     // performing nonblocking operations and spinlock until the data is available
     u32* cmd_buffer = Kernel::GetCommandBuffer();
     u32 socket_handle = cmd_buffer[1];
-    socklen_t len = cmd_buffer[2];
 
     // Memory address of the ctr_input_addr structure
     VAddr ctr_input_addr_addr = cmd_buffer[6];
diff --git a/src/core/hle/svc.cpp b/src/core/hle/svc.cpp
index 2ca270de3..2b242ff98 100644
--- a/src/core/hle/svc.cpp
+++ b/src/core/hle/svc.cpp
@@ -248,6 +248,8 @@ static ResultCode SendSyncRequest(Kernel::Handle handle) {
 
     LOG_TRACE(Kernel_SVC, "called handle=0x%08X(%s)", handle, session->GetName().c_str());
 
+    Core::System::GetInstance().PrepareReschedule();
+
     // TODO(Subv): svcSendSyncRequest should put the caller thread to sleep while the server
     // responds and cause a reschedule.
     return session->SendSyncRequest();
@@ -270,27 +272,27 @@ static ResultCode WaitSynchronization1(Kernel::Handle handle, s64 nano_seconds)
     LOG_TRACE(Kernel_SVC, "called handle=0x%08X(%s:%s), nanoseconds=%lld", handle,
               object->GetTypeName().c_str(), object->GetName().c_str(), nano_seconds);
 
-    if (object->ShouldWait()) {
+    if (object->ShouldWait(thread)) {
 
         if (nano_seconds == 0)
             return ERR_SYNC_TIMEOUT;
 
+        thread->wait_objects = {object};
         object->AddWaitingThread(thread);
-        // TODO(Subv): Perform things like update the mutex lock owner's priority to
-        // prevent priority inversion. Currently this is done in Mutex::ShouldWait,
-        // but it should be moved to a function that is called from here.
-        thread->status = THREADSTATUS_WAIT_SYNCH;
+        thread->status = THREADSTATUS_WAIT_SYNCH_ANY;
 
         // Create an event to wake the thread up after the specified nanosecond delay has passed
         thread->WakeAfterDelay(nano_seconds);
 
+        Core::System::GetInstance().PrepareReschedule();
+
         // Note: The output of this SVC will be set to RESULT_SUCCESS if the thread
         // resumes due to a signal in its wait objects.
         // Otherwise we retain the default value of timeout.
         return ERR_SYNC_TIMEOUT;
     }
 
-    object->Acquire();
+    object->Acquire(thread);
 
     return RESULT_SUCCESS;
 }
@@ -324,19 +326,14 @@ static ResultCode WaitSynchronizationN(s32* out, Kernel::Handle* handles, s32 ha
         objects[i] = object;
     }
 
-    // Clear the mapping of wait object indices.
-    // We don't want any lingering state in this map.
-    // It will be repopulated later in the wait_all = false case.
-    thread->wait_objects_index.clear();
-
     if (wait_all) {
         bool all_available =
             std::all_of(objects.begin(), objects.end(),
-                        [](const ObjectPtr& object) { return !object->ShouldWait(); });
+                        [thread](const ObjectPtr& object) { return !object->ShouldWait(thread); });
         if (all_available) {
             // We can acquire all objects right now, do so.
             for (auto& object : objects)
-                object->Acquire();
+                object->Acquire(thread);
             // Note: In this case, the `out` parameter is not set,
             // and retains whatever value it had before.
             return RESULT_SUCCESS;
@@ -350,22 +347,20 @@ static ResultCode WaitSynchronizationN(s32* out, Kernel::Handle* handles, s32 ha
             return ERR_SYNC_TIMEOUT;
 
         // Put the thread to sleep
-        thread->status = THREADSTATUS_WAIT_SYNCH;
+        thread->status = THREADSTATUS_WAIT_SYNCH_ALL;
 
         // Add the thread to each of the objects' waiting threads.
         for (auto& object : objects) {
             object->AddWaitingThread(thread);
-            // TODO(Subv): Perform things like update the mutex lock owner's priority to
-            // prevent priority inversion. Currently this is done in Mutex::ShouldWait,
-            // but it should be moved to a function that is called from here.
         }
 
-        // Set the thread's waitlist to the list of objects passed to WaitSynchronizationN
         thread->wait_objects = std::move(objects);
 
         // Create an event to wake the thread up after the specified nanosecond delay has passed
         thread->WakeAfterDelay(nano_seconds);
 
+        Core::System::GetInstance().PrepareReschedule();
+
         // This value gets set to -1 by default in this case, it is not modified after this.
         *out = -1;
         // Note: The output of this SVC will be set to RESULT_SUCCESS if the thread resumes due to
@@ -373,13 +368,14 @@ static ResultCode WaitSynchronizationN(s32* out, Kernel::Handle* handles, s32 ha
         return ERR_SYNC_TIMEOUT;
     } else {
         // Find the first object that is acquirable in the provided list of objects
-        auto itr = std::find_if(objects.begin(), objects.end(),
-                                [](const ObjectPtr& object) { return !object->ShouldWait(); });
+        auto itr = std::find_if(objects.begin(), objects.end(), [thread](const ObjectPtr& object) {
+            return !object->ShouldWait(thread);
+        });
 
         if (itr != objects.end()) {
             // We found a ready object, acquire it and set the result value
             Kernel::WaitObject* object = itr->get();
-            object->Acquire();
+            object->Acquire(thread);
             *out = std::distance(objects.begin(), itr);
             return RESULT_SUCCESS;
         }
@@ -392,28 +388,24 @@ static ResultCode WaitSynchronizationN(s32* out, Kernel::Handle* handles, s32 ha
             return ERR_SYNC_TIMEOUT;
 
         // Put the thread to sleep
-        thread->status = THREADSTATUS_WAIT_SYNCH;
-
-        // Clear the thread's waitlist, we won't use it for wait_all = false
-        thread->wait_objects.clear();
+        thread->status = THREADSTATUS_WAIT_SYNCH_ANY;
 
         // Add the thread to each of the objects' waiting threads.
         for (size_t i = 0; i < objects.size(); ++i) {
             Kernel::WaitObject* object = objects[i].get();
-            // Set the index of this object in the mapping of Objects -> index for this thread.
-            thread->wait_objects_index[object->GetObjectId()] = static_cast<int>(i);
             object->AddWaitingThread(thread);
-            // TODO(Subv): Perform things like update the mutex lock owner's priority to
-            // prevent priority inversion. Currently this is done in Mutex::ShouldWait,
-            // but it should be moved to a function that is called from here.
         }
 
+        thread->wait_objects = std::move(objects);
+
         // Note: If no handles and no timeout were given, then the thread will deadlock, this is
         // consistent with hardware behavior.
 
         // Create an event to wake the thread up after the specified nanosecond delay has passed
         thread->WakeAfterDelay(nano_seconds);
 
+        Core::System::GetInstance().PrepareReschedule();
+
         // Note: The output of this SVC will be set to RESULT_SUCCESS if the thread resumes due to a
         // signal in one of its wait objects.
         // Otherwise we retain the default value of timeout, and -1 in the out parameter
@@ -448,6 +440,9 @@ static ResultCode ArbitrateAddress(Kernel::Handle handle, u32 address, u32 type,
     auto res = arbiter->ArbitrateAddress(static_cast<Kernel::ArbitrationType>(type), address, value,
                                          nanoseconds);
 
+    // TODO(Subv): Identify in which specific cases this call should cause a reschedule.
+    Core::System::GetInstance().PrepareReschedule();
+
     return res;
 }
 
@@ -574,6 +569,8 @@ static ResultCode CreateThread(Kernel::Handle* out_handle, s32 priority, u32 ent
 
     CASCADE_RESULT(*out_handle, Kernel::g_handle_table.Create(std::move(thread)));
 
+    Core::System::GetInstance().PrepareReschedule();
+
     LOG_TRACE(Kernel_SVC, "called entrypoint=0x%08X (%s), arg=0x%08X, stacktop=0x%08X, "
                           "threadpriority=0x%08X, processorid=0x%08X : created handle=0x%08X",
               entry_point, name.c_str(), arg, stack_top, priority, processor_id, *out_handle);
@@ -586,6 +583,7 @@ static void ExitThread() {
     LOG_TRACE(Kernel_SVC, "called, pc=0x%08X", Core::CPU().GetPC());
 
     Kernel::ExitCurrentThread();
+    Core::System::GetInstance().PrepareReschedule();
 }
 
 /// Gets the priority for the specified thread
@@ -605,6 +603,13 @@ static ResultCode SetThreadPriority(Kernel::Handle handle, s32 priority) {
         return ERR_INVALID_HANDLE;
 
     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;
 }
 
@@ -844,11 +849,18 @@ static ResultCode CancelTimer(Kernel::Handle handle) {
 static void SleepThread(s64 nanoseconds) {
     LOG_TRACE(Kernel_SVC, "called nanoseconds=%lld", 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.
+    if (nanoseconds == 0 && !Kernel::HaveReadyThreads())
+        return;
+
     // Sleep current thread and check for next thread to schedule
     Kernel::WaitCurrentThread_Sleep();
 
     // Create an event to wake the thread up after the specified nanosecond delay has passed
     Kernel::GetCurrentThread()->WakeAfterDelay(nanoseconds);
+
+    Core::System::GetInstance().PrepareReschedule();
 }
 
 /// This returns the total CPU ticks elapsed since the CPU was powered-on
@@ -890,7 +902,11 @@ static ResultCode CreateMemoryBlock(Kernel::Handle* out_handle, u32 addr, u32 si
         return ResultCode(ErrorDescription::InvalidCombination, ErrorModule::OS,
                           ErrorSummary::InvalidArgument, ErrorLevel::Usage);
 
-    if (addr < Memory::PROCESS_IMAGE_VADDR || addr + size > Memory::SHARED_MEMORY_VADDR_END) {
+    // TODO(Subv): Processes with memory type APPLICATION are not allowed
+    // to create memory blocks with addr = 0, any attempts to do so
+    // should return error 0xD92007EA.
+    if ((addr < Memory::PROCESS_IMAGE_VADDR || addr + size > Memory::SHARED_MEMORY_VADDR_END) &&
+        addr != 0) {
         return ResultCode(ErrorDescription::InvalidAddress, ErrorModule::OS,
                           ErrorSummary::InvalidArgument, ErrorLevel::Usage);
     }
@@ -1184,8 +1200,6 @@ void CallSVC(u32 immediate) {
     if (info) {
         if (info->func) {
             info->func();
-            //  TODO(Subv): Not all service functions should cause a reschedule in all cases.
-            Core::System::GetInstance().PrepareReschedule();
         } else {
             LOG_ERROR(Kernel_SVC, "unimplemented SVC function %s(..)", info->name);
         }
diff --git a/src/core/loader/ncch.cpp b/src/core/loader/ncch.cpp
index a204dc336..5df33f6d2 100644
--- a/src/core/loader/ncch.cpp
+++ b/src/core/loader/ncch.cpp
@@ -288,7 +288,7 @@ ResultStatus AppLoader_NCCH::LoadExeFS() {
     LOG_DEBUG(Loader, "Thread priority:             0x%X", priority);
     LOG_DEBUG(Loader, "Resource limit category:     %d", resource_limit_category);
     LOG_DEBUG(Loader, "System Mode:                 %d",
-              exheader_header.arm11_system_local_caps.system_mode);
+              static_cast<int>(exheader_header.arm11_system_local_caps.system_mode));
 
     if (exheader_header.arm11_system_local_caps.program_id != ncch_header.program_id) {
         LOG_ERROR(Loader, "ExHeader Program ID mismatch: the ROM is probably encrypted.");
diff --git a/src/core/loader/ncch.h b/src/core/loader/ncch.h
index fe08f5b45..4ef95b5c6 100644
--- a/src/core/loader/ncch.h
+++ b/src/core/loader/ncch.h
@@ -181,7 +181,7 @@ public:
      * Loads the Exheader and returns the system mode for this application.
      * @return Optional with the kernel system mode
      */
-    boost::optional<u32> LoadKernelSystemMode();
+    boost::optional<u32> LoadKernelSystemMode() override;
 
     ResultStatus ReadCode(std::vector<u8>& buffer) override;
 
diff --git a/src/core/settings.cpp b/src/core/settings.cpp
index 5d23c52f9..9afaf79ec 100644
--- a/src/core/settings.cpp
+++ b/src/core/settings.cpp
@@ -20,7 +20,6 @@ void Apply() {
 
     VideoCore::g_hw_renderer_enabled = values.use_hw_renderer;
     VideoCore::g_shader_jit_enabled = values.use_shader_jit;
-    VideoCore::g_scaled_resolution_enabled = values.use_scaled_resolution;
     VideoCore::g_toggle_framelimit_enabled = values.toggle_framelimit;
 
     if (VideoCore::g_emu_window) {
diff --git a/src/core/settings.h b/src/core/settings.h
index 4e7a4b1be..8dbda653a 100644
--- a/src/core/settings.h
+++ b/src/core/settings.h
@@ -88,7 +88,7 @@ struct Values {
     // Renderer
     bool use_hw_renderer;
     bool use_shader_jit;
-    bool use_scaled_resolution;
+    float resolution_factor;
     bool use_vsync;
     bool toggle_framelimit;
 
diff --git a/src/video_core/renderer_opengl/gl_rasterizer_cache.cpp b/src/video_core/renderer_opengl/gl_rasterizer_cache.cpp
index 85aa06cd5..ef3b06a7b 100644
--- a/src/video_core/renderer_opengl/gl_rasterizer_cache.cpp
+++ b/src/video_core/renderer_opengl/gl_rasterizer_cache.cpp
@@ -556,14 +556,21 @@ RasterizerCacheOpenGL::GetFramebufferSurfaces(const Pica::Regs::FramebufferConfi
     color_params.width = depth_params.width = config.GetWidth();
     color_params.height = depth_params.height = config.GetHeight();
     color_params.is_tiled = depth_params.is_tiled = true;
-    if (VideoCore::g_scaled_resolution_enabled) {
-        auto layout = VideoCore::g_emu_window->GetFramebufferLayout();
 
-        // Assume same scaling factor for top and bottom screens
+    // Set the internal resolution, assume the same scaling factor for top and bottom screens
+    const Layout::FramebufferLayout& layout = VideoCore::g_emu_window->GetFramebufferLayout();
+    if (Settings::values.resolution_factor == 0.0f) {
+        // Auto - scale resolution to the window size
         color_params.res_scale_width = depth_params.res_scale_width =
             (float)layout.top_screen.GetWidth() / VideoCore::kScreenTopWidth;
         color_params.res_scale_height = depth_params.res_scale_height =
             (float)layout.top_screen.GetHeight() / VideoCore::kScreenTopHeight;
+    } else {
+        // Otherwise, scale the resolution by the specified factor
+        color_params.res_scale_width = Settings::values.resolution_factor;
+        depth_params.res_scale_width = Settings::values.resolution_factor;
+        color_params.res_scale_height = Settings::values.resolution_factor;
+        depth_params.res_scale_height = Settings::values.resolution_factor;
     }
 
     color_params.addr = config.GetColorBufferPhysicalAddress();
diff --git a/src/video_core/shader/shader_interpreter.cpp b/src/video_core/shader/shader_interpreter.cpp
index 70db4167e..20fb9754b 100644
--- a/src/video_core/shader/shader_interpreter.cpp
+++ b/src/video_core/shader/shader_interpreter.cpp
@@ -27,8 +27,6 @@ namespace Pica {
 
 namespace Shader {
 
-constexpr u32 INVALID_ADDRESS = 0xFFFFFFFF;
-
 struct CallStackElement {
     u32 final_address;  // Address upon which we jump to return_address
     u32 return_address; // Where to jump when leaving scope
diff --git a/src/video_core/video_core.cpp b/src/video_core/video_core.cpp
index 8db882f59..7186a7652 100644
--- a/src/video_core/video_core.cpp
+++ b/src/video_core/video_core.cpp
@@ -19,7 +19,6 @@ std::unique_ptr<RendererBase> g_renderer; ///< Renderer plugin
 
 std::atomic<bool> g_hw_renderer_enabled;
 std::atomic<bool> g_shader_jit_enabled;
-std::atomic<bool> g_scaled_resolution_enabled;
 std::atomic<bool> g_vsync_enabled;
 std::atomic<bool> g_toggle_framelimit_enabled;
 
diff --git a/src/video_core/video_core.h b/src/video_core/video_core.h
index c397c1974..4aba19ca0 100644
--- a/src/video_core/video_core.h
+++ b/src/video_core/video_core.h
@@ -37,7 +37,6 @@ extern EmuWindow* g_emu_window;                  ///< Emu window
 // qt ui)
 extern std::atomic<bool> g_hw_renderer_enabled;
 extern std::atomic<bool> g_shader_jit_enabled;
-extern std::atomic<bool> g_scaled_resolution_enabled;
 extern std::atomic<bool> g_toggle_framelimit_enabled;
 
 /// Start the video core