core: Use a raw pointer in GetGPUDebugContext.

This helper is called very often. The memory ownership shall not be transfered, so just return the raw pointer.

.gitignore

@@ -11,7 +11,7 @@ src/common/scm_rev.cpp
 .idea/
 .vs/
 .vscode/
-CMakeLists.txt.user
+CMakeLists.txt.user*
 
 # *nix related
 # Common convention for backup or temporary files

.gitmodules

@@ -7,6 +7,9 @@
 [submodule "catch"]
     path = externals/catch
     url = https://github.com/philsquared/Catch.git
+[submodule "cubeb"]
+    path = externals/cubeb
+    url = https://github.com/kinetiknz/cubeb.git
 [submodule "dynarmic"]
     path = externals/dynarmic
     url = https://github.com/MerryMage/dynarmic.git
@@ -22,3 +25,9 @@
 [submodule "unicorn"]
     path = externals/unicorn
     url = https://github.com/yuzu-emu/unicorn
+[submodule "mbedtls"]
+    path = externals/mbedtls
+    url = https://github.com/DarkLordZach/mbedtls
+[submodule "opus"]
+    path = externals/opus
+    url = https://github.com/ogniK5377/opus.git

.travis.yml

@@ -20,6 +20,7 @@ matrix:
       install: "./.travis/linux/deps.sh"
       script: "./.travis/linux/build.sh"
       after_success: "./.travis/linux/upload.sh"
+      cache: ccache
     - os: osx
       env: NAME="macos build"
       sudo: false
@@ -27,6 +28,7 @@ matrix:
       install: "./.travis/macos/deps.sh"
       script: "./.travis/macos/build.sh"
       after_success: "./.travis/macos/upload.sh"
+      cache: ccache
 
 deploy:
   provider: releases
@@ -42,7 +44,3 @@ notifications:
   webhooks:
     urls:
       - https://api.yuzu-emu.org/code/travis/notify
-
-cache:
-  directories:
-    - $HOME/.ccache

.travis/common/travis-ci.env

@@ -0,0 +1,12 @@
+# List of environment variables to be shared with Docker containers
+CI
+TRAVIS
+CONTINUOUS_INTEGRATION
+TRAVIS_BRANCH
+TRAVIS_BUILD_ID
+TRAVIS_BUILD_NUMBER
+TRAVIS_COMMIT
+TRAVIS_JOB_ID
+TRAVIS_JOB_NUMBER
+TRAVIS_REPO_SLUG
+TRAVIS_TAG

.travis/linux/build.sh

@@ -1,3 +1,4 @@
 #!/bin/bash -ex
 
-docker run -e CCACHE_DIR=/ccache -v $HOME/.ccache:/ccache -v $(pwd):/yuzu ubuntu:18.04 /bin/bash /yuzu/.travis/linux/docker.sh
+mkdir -p "$HOME/.ccache"
+docker run --env-file .travis/common/travis-ci.env -v $(pwd):/yuzu -v "$HOME/.ccache":/root/.ccache ubuntu:18.04 /bin/bash /yuzu/.travis/linux/docker.sh

.travis/linux/docker.sh

@@ -5,14 +5,8 @@ apt-get install --no-install-recommends -y build-essential git libqt5opengl5-dev
 
 cd /yuzu
 
-export PATH=/usr/lib/ccache:$PATH
-ln -sf /usr/bin/ccache /usr/lib/ccache/cc
-ln -sf /usr/bin/ccache /usr/lib/ccache/c++
 mkdir build && cd build
-ccache --show-stats > ccache_before
-cmake .. -DYUZU_BUILD_UNICORN=ON -DCMAKE_BUILD_TYPE=Release -G Ninja
+cmake .. -DYUZU_BUILD_UNICORN=ON -DCMAKE_BUILD_TYPE=Release -DCMAKE_C_COMPILER=/usr/lib/ccache/gcc -DCMAKE_CXX_COMPILER=/usr/lib/ccache/g++ -DENABLE_COMPATIBILITY_LIST_DOWNLOAD=ON -G Ninja
 ninja
-ccache --show-stats > ccache_after
-diff -U100 ccache_before ccache_after || true
 
 ctest -VV -C Release

.travis/macos/build.sh

@@ -5,14 +5,11 @@ set -o pipefail
 export MACOSX_DEPLOYMENT_TARGET=10.12
 export Qt5_DIR=$(brew --prefix)/opt/qt5
 export UNICORNDIR=$(pwd)/externals/unicorn
+export PATH="/usr/local/opt/ccache/libexec:$PATH"
 
 mkdir build && cd build
-export PATH=/usr/local/opt/ccache/libexec:$PATH
-ccache --show-stats > ccache_before
 cmake --version
-cmake .. -DYUZU_BUILD_UNICORN=ON -DCMAKE_BUILD_TYPE=Release
+cmake .. -DYUZU_BUILD_UNICORN=ON -DCMAKE_BUILD_TYPE=Release -DENABLE_COMPATIBILITY_LIST_DOWNLOAD=ON
 make -j4
-ccache --show-stats > ccache_after
-diff -U100 ccache_before ccache_after || true
 
 ctest -VV -C Release

CMakeLists.txt

@@ -1,5 +1,5 @@
-# CMake 3.6 required for FindBoost to define IMPORTED libs properly on unknown Boost versions
-cmake_minimum_required(VERSION 3.6)
+cmake_minimum_required(VERSION 3.7)
+
 list(APPEND CMAKE_MODULE_PATH "${CMAKE_CURRENT_SOURCE_DIR}/CMakeModules")
 list(APPEND CMAKE_MODULE_PATH "${CMAKE_CURRENT_SOURCE_DIR}/externals/cmake-modules")
 include(DownloadExternals)
@@ -17,6 +17,8 @@ CMAKE_DEPENDENT_OPTION(YUZU_USE_BUNDLED_QT "Download bundled Qt binaries" ON "EN
 
 option(YUZU_USE_BUNDLED_UNICORN "Build/Download bundled Unicorn" ON)
 
+option(ENABLE_CUBEB "Enables the cubeb audio backend" ON)
+
 if(NOT EXISTS ${CMAKE_SOURCE_DIR}/.git/hooks/pre-commit)
     message(STATUS "Copying pre-commit hook")
     file(COPY hooks/pre-commit
@@ -39,6 +41,19 @@ function(check_submodules_present)
 endfunction()
 check_submodules_present()
 
+configure_file(${CMAKE_SOURCE_DIR}/dist/compatibility_list/compatibility_list.qrc
+               ${CMAKE_BINARY_DIR}/dist/compatibility_list/compatibility_list.qrc
+               COPYONLY)
+if (ENABLE_COMPATIBILITY_LIST_DOWNLOAD AND NOT EXISTS ${CMAKE_BINARY_DIR}/dist/compatibility_list/compatibility_list.json)
+    message(STATUS "Downloading compatibility list for yuzu...")
+    file(DOWNLOAD
+        https://api.yuzu-emu.org/gamedb/
+        "${CMAKE_BINARY_DIR}/dist/compatibility_list/compatibility_list.json" SHOW_PROGRESS)
+endif()
+if (NOT EXISTS ${CMAKE_BINARY_DIR}/dist/compatibility_list/compatibility_list.json)
+    file(WRITE ${CMAKE_BINARY_DIR}/dist/compatibility_list/compatibility_list.json "")
+endif()
+
 # Detect current compilation architecture and create standard definitions
 # =======================================================================
 
@@ -64,10 +79,12 @@ if (NOT ENABLE_GENERIC)
         detect_architecture("_M_AMD64" x86_64)
         detect_architecture("_M_IX86" x86)
         detect_architecture("_M_ARM" ARM)
+        detect_architecture("_M_ARM64" ARM64)
     else()
         detect_architecture("__x86_64__" x86_64)
         detect_architecture("__i386__" x86)
         detect_architecture("__arm__" ARM)
+        detect_architecture("__aarch64__" ARM64)
     endif()
 endif()
 
@@ -185,8 +202,8 @@ find_package(Threads REQUIRED)
 if (ENABLE_SDL2)
     if (YUZU_USE_BUNDLED_SDL2)
         # Detect toolchain and platform
-        if (MSVC14 AND ARCHITECTURE_x86_64)
-            set(SDL2_VER "SDL2-2.0.5")
+        if ((MSVC_VERSION GREATER_EQUAL 1910 AND MSVC_VERSION LESS 1920) AND ARCHITECTURE_x86_64)
+            set(SDL2_VER "SDL2-2.0.8")
         else()
             message(FATAL_ERROR "No bundled SDL2 binaries for your toolchain. Disable YUZU_USE_BUNDLED_SDL2 and provide your own.")
         endif()
@@ -218,7 +235,7 @@ if (YUZU_USE_BUNDLED_UNICORN)
     if (MSVC)
         message(STATUS "unicorn not found, falling back to bundled")
         # Detect toolchain and platform
-        if (MSVC14 AND ARCHITECTURE_x86_64)
+        if ((MSVC_VERSION GREATER_EQUAL 1910 AND MSVC_VERSION LESS 1920) AND ARCHITECTURE_x86_64)
             set(UNICORN_VER "unicorn-yuzu")
         else()
             message(FATAL_ERROR "No bundled Unicorn binaries for your toolchain. Disable YUZU_USE_BUNDLED_UNICORN and provide your own.")
@@ -277,7 +294,7 @@ endif()
 
 if (ENABLE_QT)
     if (YUZU_USE_BUNDLED_QT)
-        if (MSVC14 AND ARCHITECTURE_x86_64)
+        if ((MSVC_VERSION GREATER_EQUAL 1910 AND MSVC_VERSION LESS 1920) AND ARCHITECTURE_x86_64)
             set(QT_VER qt-5.10.0-msvc2015_64)
         else()
             message(FATAL_ERROR "No bundled Qt binaries for your toolchain. Disable YUZU_USE_BUNDLED_QT and provide your own.")
@@ -301,7 +318,7 @@ endif()
 # ======================================
 
 IF (APPLE)
-    FIND_LIBRARY(COCOA_LIBRARY Cocoa)           # Umbrella framework for everything GUI-related
+    find_library(COCOA_LIBRARY Cocoa)           # Umbrella framework for everything GUI-related
     set(PLATFORM_LIBRARIES ${COCOA_LIBRARY} ${IOKIT_LIBRARY} ${COREVIDEO_LIBRARY})
 
     if (CMAKE_CXX_COMPILER_ID STREQUAL Clang)

CMakeModules/CopyYuzuQt5Deps.cmake

@@ -4,8 +4,10 @@ function(copy_yuzu_Qt5_deps target_dir)
     set(Qt5_DLL_DIR "${Qt5_DIR}/../../../bin")
     set(Qt5_PLATFORMS_DIR "${Qt5_DIR}/../../../plugins/platforms/")
     set(Qt5_STYLES_DIR "${Qt5_DIR}/../../../plugins/styles/")
+    set(Qt5_IMAGEFORMATS_DIR "${Qt5_DIR}/../../../plugins/imageformats/")
     set(PLATFORMS ${DLL_DEST}platforms/)
     set(STYLES ${DLL_DEST}styles/)
+    set(IMAGEFORMATS ${DLL_DEST}imageformats/)
     windows_copy_files(${target_dir} ${Qt5_DLL_DIR} ${DLL_DEST}
         icudt*.dll
         icuin*.dll
@@ -17,4 +19,5 @@ function(copy_yuzu_Qt5_deps target_dir)
     )
     windows_copy_files(yuzu ${Qt5_PLATFORMS_DIR} ${PLATFORMS} qwindows$<$<CONFIG:Debug>:d>.*)
     windows_copy_files(yuzu ${Qt5_STYLES_DIR} ${STYLES} qwindowsvistastyle$<$<CONFIG:Debug>:d>.*)
+    windows_copy_files(yuzu ${Qt5_IMAGEFORMATS_DIR} ${IMAGEFORMATS} qjpeg$<$<CONFIG:Debug>:d>.*)
 endfunction(copy_yuzu_Qt5_deps)

appveyor.yml

@@ -41,9 +41,9 @@ before_build:
   - ps: |
         if ($env:BUILD_TYPE -eq 'msvc') {
           # redirect stderr and change the exit code to prevent powershell from cancelling the build if cmake prints a warning
-          cmd /C 'cmake -G "Visual Studio 15 2017 Win64" -DYUZU_USE_BUNDLED_QT=1 -DYUZU_USE_BUNDLED_SDL2=1 -DYUZU_USE_BUNDLED_UNICORN=1 .. 2>&1 && exit 0'
+          cmd /C 'cmake -G "Visual Studio 15 2017 Win64" -DYUZU_USE_BUNDLED_QT=1 -DYUZU_USE_BUNDLED_SDL2=1 -DYUZU_USE_BUNDLED_UNICORN=1 -DENABLE_COMPATIBILITY_LIST_DOWNLOAD=ON .. 2>&1 && exit 0'
         } else {
-          C:\msys64\usr\bin\bash.exe -lc "cmake -G 'MSYS Makefiles' -DYUZU_BUILD_UNICORN=1 -DCMAKE_BUILD_TYPE=Release .. 2>&1"
+          C:\msys64\usr\bin\bash.exe -lc "cmake -G 'MSYS Makefiles' -DYUZU_BUILD_UNICORN=1 -DCMAKE_BUILD_TYPE=Release -DENABLE_COMPATIBILITY_LIST_DOWNLOAD=ON .. 2>&1"
         }
   - cd ..
 
@@ -117,6 +117,7 @@ after_build:
           mkdir $RELEASE_DIST
           mkdir $RELEASE_DIST/platforms
           mkdir $RELEASE_DIST/styles
+          mkdir $RELEASE_DIST/imageformats
 
           # copy the compiled binaries and other release files to the release folder
           Get-ChildItem "$CMAKE_BINARY_DIR" -Filter "yuzu*.exe" | Copy-Item -destination $RELEASE_DIST
@@ -140,6 +141,9 @@ after_build:
           # copy the qt windows vista style dll to platforms
           Copy-Item -path "C:/msys64/mingw64/share/qt5/plugins/styles/qwindowsvistastyle.dll" -force -destination "$RELEASE_DIST/styles"
 
+          # copy the qt jpeg imageformat dll to platforms
+          Copy-Item -path "C:/msys64/mingw64/share/qt5/plugins/imageformats/qjpeg.dll" -force -destination "$RELEASE_DIST/imageformats"
+
           7z a -tzip $MINGW_BUILD_ZIP $RELEASE_DIST\*
           7z a $MINGW_SEVENZIP $RELEASE_DIST
         }
@@ -158,10 +162,6 @@ artifacts:
   - path: $(BUILD_ZIP)
     name: build
     type: zip
-  - path: $(BUILD_SYMBOLS)
-    name: debugsymbols
-  - path: $(BUILD_UPDATE)
-    name: update
 
 deploy:
   provider: GitHub

dist/compatibility_list/compatibility_list.qrc

@@ -0,0 +1,5 @@
+<RCC>
+  <qresource prefix="compatibility_list">
+      <file>compatibility_list.json</file>
+  </qresource>
+</RCC>

dist/qt_themes/qdarkstyle/LICENSE.md

@@ -0,0 +1,183 @@
+# License
+
+## The MIT License (MIT) - Code
+
+Copyright (c) 2013-2018 Colin Duquesnoy
+
+Permission is hereby granted, free of charge, to any person obtaining a copy
+of this software and associated documentation files (the "Software"), to deal
+in the Software without restriction, including without limitation the rights
+to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
+copies of the Software, and to permit persons to whom the Software is
+furnished to do so, subject to the following conditions:
+
+The above copyright notice and this permission notice shall be included in
+all copies or substantial portions of the Software.
+
+THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
+IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
+FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
+AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
+LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
+OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN
+THE SOFTWARE.
+
+## Creative Commons Attribution International 4.0 - Images
+
+QDarkStyle (c) 2013-2018 Colin Duquesnoy
+
+Creative Commons Corporation (“Creative Commons”) is not a law firm and does not provide legal services or legal advice. Distribution of Creative Commons public licenses does not create a lawyer-client or other relationship. Creative Commons makes its licenses and related information available on an “as-is” basis. Creative Commons gives no warranties regarding its licenses, any material licensed under their terms and conditions, or any related information. Creative Commons disclaims all liability for damages resulting from their use to the fullest extent possible.
+
+### Using Creative Commons Public Licenses
+
+Creative Commons public licenses provide a standard set of terms and conditions that creators and other rights holders may use to share original works of authorship and other material subject to copyright and certain other rights specified in the public license below. The following considerations are for informational purposes only, are not exhaustive, and do not form part of our licenses.
+
+* __Considerations for licensors:__ Our public licenses are intended for use by those authorized to give the public permission to use material in ways otherwise restricted by copyright and certain other rights. Our licenses are irrevocable. Licensors should read and understand the terms and conditions of the license they choose before applying it. Licensors should also secure all rights necessary before applying our licenses so that the public can reuse the material as expected. Licensors should clearly mark any material not subject to the license. This includes other CC-licensed material, or material used under an exception or limitation to copyright. [More considerations for licensors](http://wiki.creativecommons.org/Considerations_for_licensors_and_licensees#Considerations_for_licensors).
+
+* __Considerations for the public:__ By using one of our public licenses, a licensor grants the public permission to use the licensed material under specified terms and conditions. If the licensor’s permission is not necessary for any reason–for example, because of any applicable exception or limitation to copyright–then that use is not regulated by the license. Our licenses grant only permissions under copyright and certain other rights that a licensor has authority to grant. Use of the licensed material may still be restricted for other reasons, including because others have copyright or other rights in the material. A licensor may make special requests, such as asking that all changes be marked or described. Although not required by our licenses, you are encouraged to respect those requests where reasonable. [More considerations for the public](http://wiki.creativecommons.org/Considerations_for_licensors_and_licensees#Considerations_for_licensees).
+
+## Creative Commons Attribution 4.0 International Public License
+
+By exercising the Licensed Rights (defined below), You accept and agree to be bound by the terms and conditions of this Creative Commons Attribution 4.0 International Public License ("Public License"). To the extent this Public License may be interpreted as a contract, You are granted the Licensed Rights in consideration of Your acceptance of these terms and conditions, and the Licensor grants You such rights in consideration of benefits the Licensor receives from making the Licensed Material available under these terms and conditions.
+
+### Section 1 – Definitions
+
+a. __Adapted Material__ means material subject to Copyright and Similar Rights that is derived from or based upon the Licensed Material and in which the Licensed Material is translated, altered, arranged, transformed, or otherwise modified in a manner requiring permission under the Copyright and Similar Rights held by the Licensor. For purposes of this Public License, where the Licensed Material is a musical work, performance, or sound recording, Adapted Material is always produced where the Licensed Material is synched in timed relation with a moving image.
+
+b. __Adapter's License__ means the license You apply to Your Copyright and Similar Rights in Your contributions to Adapted Material in accordance with the terms and conditions of this Public License.
+
+c. __Copyright and Similar Rights__ means copyright and/or similar rights closely related to copyright including, without limitation, performance, broadcast, sound recording, and Sui Generis Database Rights, without regard to how the rights are labeled or categorized. For purposes of this Public License, the rights specified in Section 2(b)(1)-(2) are not Copyright and Similar Rights.
+
+d. __Effective Technological Measures__ means those measures that, in the absence of proper authority, may not be circumvented under laws fulfilling obligations under Article 11 of the WIPO Copyright Treaty adopted on December 20, 1996, and/or similar international agreements.
+
+e. __Exceptions and Limitations__ means fair use, fair dealing, and/or any other exception or limitation to Copyright and Similar Rights that applies to Your use of the Licensed Material.
+
+f. __Licensed Material__ means the artistic or literary work, database, or other material to which the Licensor applied this Public License.
+
+g. __Licensed Rights__ means the rights granted to You subject to the terms and conditions of this Public License, which are limited to all Copyright and Similar Rights that apply to Your use of the Licensed Material and that the Licensor has authority to license.
+
+h. __Licensor__ means the individual(s) or entity(ies) granting rights under this Public License.
+
+i. __Share__ means to provide material to the public by any means or process that requires permission under the Licensed Rights, such as reproduction, public display, public performance, distribution, dissemination, communication, or importation, and to make material available to the public including in ways that members of the public may access the material from a place and at a time individually chosen by them.
+
+j. __Sui Generis Database Rights__ means rights other than copyright resulting from Directive 96/9/EC of the European Parliament and of the Council of 11 March 1996 on the legal protection of databases, as amended and/or succeeded, as well as other essentially equivalent rights anywhere in the world.
+
+k. __You__ means the individual or entity exercising the Licensed Rights under this Public License. Your has a corresponding meaning.
+
+### Section 2 – Scope
+
+a. ___License grant.___
+
+   1. Subject to the terms and conditions of this Public License, the Licensor hereby grants You a worldwide, royalty-free, non-sublicensable, non-exclusive, irrevocable license to exercise the Licensed Rights in the Licensed Material to:
+
+       A. reproduce and Share the Licensed Material, in whole or in part; and
+
+       B. produce, reproduce, and Share Adapted Material.
+
+   2. __Exceptions and Limitations.__ For the avoidance of doubt, where Exceptions and Limitations apply to Your use, this Public License does not apply, and You do not need to comply with its terms and conditions.
+
+   3. __Term.__ The term of this Public License is specified in Section 6(a).
+
+   4. __Media and formats; technical modifications allowed.__ The Licensor authorizes You to exercise the Licensed Rights in all media and formats whether now known or hereafter created, and to make technical modifications necessary to do so. The Licensor waives and/or agrees not to assert any right or authority to forbid You from making technical modifications necessary to exercise the Licensed Rights, including technical modifications necessary to circumvent Effective Technological Measures. For purposes of this Public License, simply making modifications authorized by this Section 2(a)(4) never produces Adapted Material.
+
+   5. __Downstream recipients.__
+
+       A. __Offer from the Licensor – Licensed Material.__ Every recipient of the Licensed Material automatically receives an offer from the Licensor to exercise the Licensed Rights under the terms and conditions of this Public License.
+
+       B. __No downstream restrictions.__ You may not offer or impose any additional or different terms or conditions on, or apply any Effective Technological Measures to, the Licensed Material if doing so restricts exercise of the Licensed Rights by any recipient of the Licensed Material.
+
+   6. __No endorsement.__ Nothing in this Public License constitutes or may be construed as permission to assert or imply that You are, or that Your use of the Licensed Material is, connected with, or sponsored, endorsed, or granted official status by, the Licensor or others designated to receive attribution as provided in Section 3(a)(1)(A)(i).
+
+b. ___Other rights.___
+
+   1. Moral rights, such as the right of integrity, are not licensed under this Public License, nor are publicity, privacy, and/or other similar personality rights; however, to the extent possible, the Licensor waives and/or agrees not to assert any such rights held by the Licensor to the limited extent necessary to allow You to exercise the Licensed Rights, but not otherwise.
+
+   2. Patent and trademark rights are not licensed under this Public License.
+
+   3. To the extent possible, the Licensor waives any right to collect royalties from You for the exercise of the Licensed Rights, whether directly or through a collecting society under any voluntary or waivable statutory or compulsory licensing scheme. In all other cases the Licensor expressly reserves any right to collect such royalties.
+
+### Section 3 – License Conditions
+
+Your exercise of the Licensed Rights is expressly made subject to the following conditions.
+
+a. ___Attribution.___
+
+   1. If You Share the Licensed Material (including in modified form), You must:
+
+       A. retain the following if it is supplied by the Licensor with the Licensed Material:
+
+         i. identification of the creator(s) of the Licensed Material and any others designated to receive attribution, in any reasonable manner requested by the Licensor (including by pseudonym if designated);
+
+         ii. a copyright notice;
+
+         iii. a notice that refers to this Public License;
+
+         iv. a notice that refers to the disclaimer of warranties;
+
+         v. a URI or hyperlink to the Licensed Material to the extent reasonably practicable;
+
+       B. indicate if You modified the Licensed Material and retain an indication of any previous modifications; and
+
+       C. indicate the Licensed Material is licensed under this Public License, and include the text of, or the URI or hyperlink to, this Public License.
+
+   2. You may satisfy the conditions in Section 3(a)(1) in any reasonable manner based on the medium, means, and context in which You Share the Licensed Material. For example, it may be reasonable to satisfy the conditions by providing a URI or hyperlink to a resource that includes the required information.
+
+   3. If requested by the Licensor, You must remove any of the information required by Section 3(a)(1)(A) to the extent reasonably practicable.
+
+   4. If You Share Adapted Material You produce, the Adapter's License You apply must not prevent recipients of the Adapted Material from complying with this Public License.
+
+### Section 4 – Sui Generis Database Rights
+
+Where the Licensed Rights include Sui Generis Database Rights that apply to Your use of the Licensed Material:
+
+a. for the avoidance of doubt, Section 2(a)(1) grants You the right to extract, reuse, reproduce, and Share all or a substantial portion of the contents of the database;
+
+b. if You include all or a substantial portion of the database contents in a database in which You have Sui Generis Database Rights, then the database in which You have Sui Generis Database Rights (but not its individual contents) is Adapted Material; and
+
+c. You must comply with the conditions in Section 3(a) if You Share all or a substantial portion of the contents of the database.
+
+For the avoidance of doubt, this Section 4 supplements and does not replace Your obligations under this Public License where the Licensed Rights include other Copyright and Similar Rights.
+
+### Section 5 – Disclaimer of Warranties and Limitation of Liability
+
+a. __Unless otherwise separately undertaken by the Licensor, to the extent possible, the Licensor offers the Licensed Material as-is and as-available, and makes no representations or warranties of any kind concerning the Licensed Material, whether express, implied, statutory, or other. This includes, without limitation, warranties of title, merchantability, fitness for a particular purpose, non-infringement, absence of latent or other defects, accuracy, or the presence or absence of errors, whether or not known or discoverable. Where disclaimers of warranties are not allowed in full or in part, this disclaimer may not apply to You.__
+
+b. __To the extent possible, in no event will the Licensor be liable to You on any legal theory (including, without limitation, negligence) or otherwise for any direct, special, indirect, incidental, consequential, punitive, exemplary, or other losses, costs, expenses, or damages arising out of this Public License or use of the Licensed Material, even if the Licensor has been advised of the possibility of such losses, costs, expenses, or damages. Where a limitation of liability is not allowed in full or in part, this limitation may not apply to You.__
+
+c. The disclaimer of warranties and limitation of liability provided above shall be interpreted in a manner that, to the extent possible, most closely approximates an absolute disclaimer and waiver of all liability.
+
+### Section 6 – Term and Termination
+
+a. This Public License applies for the term of the Copyright and Similar Rights licensed here. However, if You fail to comply with this Public License, then Your rights under this Public License terminate automatically.
+
+b. Where Your right to use the Licensed Material has terminated under Section 6(a), it reinstates:
+
+   1. automatically as of the date the violation is cured, provided it is cured within 30 days of Your discovery of the violation; or
+
+   2. upon express reinstatement by the Licensor.
+
+   For the avoidance of doubt, this Section 6(b) does not affect any right the Licensor may have to seek remedies for Your violations of this Public License.
+
+c. For the avoidance of doubt, the Licensor may also offer the Licensed Material under separate terms or conditions or stop distributing the Licensed Material at any time; however, doing so will not terminate this Public License.
+
+d. Sections 1, 5, 6, 7, and 8 survive termination of this Public License.
+
+### Section 7 – Other Terms and Conditions
+
+a. The Licensor shall not be bound by any additional or different terms or conditions communicated by You unless expressly agreed.
+
+b. Any arrangements, understandings, or agreements regarding the Licensed Material not stated herein are separate from and independent of the terms and conditions of this Public License.
+
+### Section 8 – Interpretation
+
+a. For the avoidance of doubt, this Public License does not, and shall not be interpreted to, reduce, limit, restrict, or impose conditions on any use of the Licensed Material that could lawfully be made without permission under this Public License.
+
+b. To the extent possible, if any provision of this Public License is deemed unenforceable, it shall be automatically reformed to the minimum extent necessary to make it enforceable. If the provision cannot be reformed, it shall be severed from this Public License without affecting the enforceability of the remaining terms and conditions.
+
+c. No term or condition of this Public License will be waived and no failure to comply consented to unless expressly agreed to by the Licensor.
+
+d. Nothing in this Public License constitutes or may be interpreted as a limitation upon, or waiver of, any privileges and immunities that apply to the Licensor or You, including from the legal processes of any jurisdiction or authority.
+
+> Creative Commons is not a party to its public licenses. Notwithstanding, Creative Commons may elect to apply one of its public licenses to material it publishes and in those instances will be considered the “Licensor.” Except for the limited purpose of indicating that material is shared under a Creative Commons public license or as otherwise permitted by the Creative Commons policies published at [creativecommons.org/policies](http://creativecommons.org/policies), Creative Commons does not authorize the use of the trademark “Creative Commons” or any other trademark or logo of Creative Commons without its prior written consent including, without limitation, in connection with any unauthorized modifications to any of its public licenses or any other arrangements, understandings, or agreements concerning use of licensed material. For the avoidance of doubt, this paragraph does not form part of the public licenses.
+>
+> Creative Commons may be contacted at creativecommons.org

externals/CMakeLists.txt

@@ -32,9 +32,13 @@ add_subdirectory(inih)
 
 # lz4
 set(LZ4_BUNDLED_MODE ON)
-add_subdirectory(lz4/contrib/cmake_unofficial)
+add_subdirectory(lz4/contrib/cmake_unofficial EXCLUDE_FROM_ALL)
 target_include_directories(lz4_static INTERFACE ./lz4/lib)
 
+# mbedtls
+add_subdirectory(mbedtls EXCLUDE_FROM_ALL)
+target_include_directories(mbedtls PUBLIC ./mbedtls/include)
+
 # MicroProfile
 add_library(microprofile INTERFACE)
 target_include_directories(microprofile INTERFACE ./microprofile)
@@ -50,3 +54,13 @@ if (ARCHITECTURE_x86_64)
     target_include_directories(xbyak INTERFACE ./xbyak/xbyak)
     target_compile_definitions(xbyak INTERFACE XBYAK_NO_OP_NAMES)
 endif()
+
+# Opus
+add_subdirectory(opus)
+target_include_directories(opus INTERFACE ./opus/include)
+
+# Cubeb
+if(ENABLE_CUBEB)
+    set(BUILD_TESTS OFF CACHE BOOL "")
+    add_subdirectory(cubeb EXCLUDE_FROM_ALL)
+endif()

src/CMakeLists.txt

@@ -3,6 +3,7 @@ include_directories(.)
 
 add_subdirectory(common)
 add_subdirectory(core)
+add_subdirectory(audio_core)
 add_subdirectory(video_core)
 add_subdirectory(input_common)
 add_subdirectory(tests)

src/audio_core/CMakeLists.txt

@@ -0,0 +1,31 @@
+add_library(audio_core STATIC
+    algorithm/filter.cpp
+    algorithm/filter.h
+    algorithm/interpolate.cpp
+    algorithm/interpolate.h
+    audio_out.cpp
+    audio_out.h
+    audio_renderer.cpp
+    audio_renderer.h
+    buffer.h
+    codec.cpp
+    codec.h
+    null_sink.h
+    sink.h
+    sink_details.cpp
+    sink_details.h
+    sink_stream.h
+    stream.cpp
+    stream.h
+
+    $<$<BOOL:${ENABLE_CUBEB}>:cubeb_sink.cpp cubeb_sink.h>
+)
+
+create_target_directory_groups(audio_core)
+
+target_link_libraries(audio_core PUBLIC common core)
+
+if(ENABLE_CUBEB)
+    target_link_libraries(audio_core PRIVATE cubeb)
+    target_compile_definitions(audio_core PRIVATE -DHAVE_CUBEB=1)
+endif()

src/audio_core/algorithm/filter.cpp

@@ -0,0 +1,79 @@
+// Copyright 2018 yuzu Emulator Project
+// Licensed under GPLv2 or any later version
+// Refer to the license.txt file included.
+
+#define _USE_MATH_DEFINES
+
+#include <algorithm>
+#include <array>
+#include <cmath>
+#include <vector>
+#include "audio_core/algorithm/filter.h"
+#include "common/common_types.h"
+
+namespace AudioCore {
+
+Filter Filter::LowPass(double cutoff, double Q) {
+    const double w0 = 2.0 * M_PI * cutoff;
+    const double sin_w0 = std::sin(w0);
+    const double cos_w0 = std::cos(w0);
+    const double alpha = sin_w0 / (2 * Q);
+
+    const double a0 = 1 + alpha;
+    const double a1 = -2.0 * cos_w0;
+    const double a2 = 1 - alpha;
+    const double b0 = 0.5 * (1 - cos_w0);
+    const double b1 = 1.0 * (1 - cos_w0);
+    const double b2 = 0.5 * (1 - cos_w0);
+
+    return {a0, a1, a2, b0, b1, b2};
+}
+
+Filter::Filter() : Filter(1.0, 0.0, 0.0, 1.0, 0.0, 0.0) {}
+
+Filter::Filter(double a0, double a1, double a2, double b0, double b1, double b2)
+    : a1(a1 / a0), a2(a2 / a0), b0(b0 / a0), b1(b1 / a0), b2(b2 / a0) {}
+
+void Filter::Process(std::vector<s16>& signal) {
+    const size_t num_frames = signal.size() / 2;
+    for (size_t i = 0; i < num_frames; i++) {
+        std::rotate(in.begin(), in.end() - 1, in.end());
+        std::rotate(out.begin(), out.end() - 1, out.end());
+
+        for (size_t ch = 0; ch < channel_count; ch++) {
+            in[0][ch] = signal[i * channel_count + ch];
+
+            out[0][ch] = b0 * in[0][ch] + b1 * in[1][ch] + b2 * in[2][ch] - a1 * out[1][ch] -
+                         a2 * out[2][ch];
+
+            signal[i * 2 + ch] = static_cast<s16>(std::clamp(out[0][ch], -32768.0, 32767.0));
+        }
+    }
+}
+
+/// Calculates the appropriate Q for each biquad in a cascading filter.
+/// @param total_count The total number of biquads to be cascaded.
+/// @param index 0-index of the biquad to calculate the Q value for.
+static double CascadingBiquadQ(size_t total_count, size_t index) {
+    const double pole = M_PI * (2 * index + 1) / (4.0 * total_count);
+    return 1.0 / (2.0 * std::cos(pole));
+}
+
+CascadingFilter CascadingFilter::LowPass(double cutoff, size_t cascade_size) {
+    std::vector<Filter> cascade(cascade_size);
+    for (size_t i = 0; i < cascade_size; i++) {
+        cascade[i] = Filter::LowPass(cutoff, CascadingBiquadQ(cascade_size, i));
+    }
+    return CascadingFilter{std::move(cascade)};
+}
+
+CascadingFilter::CascadingFilter() = default;
+CascadingFilter::CascadingFilter(std::vector<Filter> filters) : filters(std::move(filters)) {}
+
+void CascadingFilter::Process(std::vector<s16>& signal) {
+    for (auto& filter : filters) {
+        filter.Process(signal);
+    }
+}
+
+} // namespace AudioCore

src/audio_core/algorithm/filter.h

@@ -0,0 +1,62 @@
+// Copyright 2018 yuzu Emulator Project
+// Licensed under GPLv2 or any later version
+// Refer to the license.txt file included.
+
+#pragma once
+
+#include <array>
+#include <vector>
+#include "common/common_types.h"
+
+namespace AudioCore {
+
+/// Digital biquad filter:
+///
+///          b0 + b1 z^-1 + b2 z^-2
+///  H(z) = ------------------------
+///          a0 + a1 z^-1 + b2 z^-2
+class Filter {
+public:
+    /// Creates a low-pass filter.
+    /// @param cutoff Determines the cutoff frequency. A value from 0.0 to 1.0.
+    /// @param Q Determines the quality factor of this filter.
+    static Filter LowPass(double cutoff, double Q = 0.7071);
+
+    /// Passthrough filter.
+    Filter();
+
+    Filter(double a0, double a1, double a2, double b0, double b1, double b2);
+
+    void Process(std::vector<s16>& signal);
+
+private:
+    static constexpr size_t channel_count = 2;
+
+    /// Coefficients are in normalized form (a0 = 1.0).
+    double a1, a2, b0, b1, b2;
+    /// Input History
+    std::array<std::array<double, channel_count>, 3> in;
+    /// Output History
+    std::array<std::array<double, channel_count>, 3> out;
+};
+
+/// Cascade filters to build up higher-order filters from lower-order ones.
+class CascadingFilter {
+public:
+    /// Creates a cascading low-pass filter.
+    /// @param cutoff Determines the cutoff frequency. A value from 0.0 to 1.0.
+    /// @param cascade_size Number of biquads in cascade.
+    static CascadingFilter LowPass(double cutoff, size_t cascade_size);
+
+    /// Passthrough.
+    CascadingFilter();
+
+    explicit CascadingFilter(std::vector<Filter> filters);
+
+    void Process(std::vector<s16>& signal);
+
+private:
+    std::vector<Filter> filters;
+};
+
+} // namespace AudioCore

src/audio_core/algorithm/interpolate.cpp

@@ -0,0 +1,71 @@
+// Copyright 2018 yuzu Emulator Project
+// Licensed under GPLv2 or any later version
+// Refer to the license.txt file included.
+
+#define _USE_MATH_DEFINES
+
+#include <algorithm>
+#include <cmath>
+#include <vector>
+#include "audio_core/algorithm/interpolate.h"
+#include "common/common_types.h"
+#include "common/logging/log.h"
+
+namespace AudioCore {
+
+/// The Lanczos kernel
+static double Lanczos(size_t a, double x) {
+    if (x == 0.0)
+        return 1.0;
+    const double px = M_PI * x;
+    return a * std::sin(px) * std::sin(px / a) / (px * px);
+}
+
+std::vector<s16> Interpolate(InterpolationState& state, std::vector<s16> input, double ratio) {
+    if (input.size() < 2)
+        return {};
+
+    if (ratio <= 0) {
+        LOG_CRITICAL(Audio, "Nonsensical interpolation ratio {}", ratio);
+        ratio = 1.0;
+    }
+
+    if (ratio != state.current_ratio) {
+        const double cutoff_frequency = std::min(0.5 / ratio, 0.5 * ratio);
+        state.nyquist = CascadingFilter::LowPass(std::clamp(cutoff_frequency, 0.0, 0.4), 3);
+        state.current_ratio = ratio;
+    }
+    state.nyquist.Process(input);
+
+    constexpr size_t taps = InterpolationState::lanczos_taps;
+    const size_t num_frames = input.size() / 2;
+
+    std::vector<s16> output;
+    output.reserve(static_cast<size_t>(input.size() / ratio + 4));
+
+    double& pos = state.position;
+    auto& h = state.history;
+    for (size_t i = 0; i < num_frames; ++i) {
+        std::rotate(h.begin(), h.end() - 1, h.end());
+        h[0][0] = input[i * 2 + 0];
+        h[0][1] = input[i * 2 + 1];
+
+        while (pos <= 1.0) {
+            double l = 0.0;
+            double r = 0.0;
+            for (size_t j = 0; j < h.size(); j++) {
+                l += Lanczos(taps, pos + j - taps + 1) * h[j][0];
+                r += Lanczos(taps, pos + j - taps + 1) * h[j][1];
+            }
+            output.emplace_back(static_cast<s16>(std::clamp(l, -32768.0, 32767.0)));
+            output.emplace_back(static_cast<s16>(std::clamp(r, -32768.0, 32767.0)));
+
+            pos += ratio;
+        }
+        pos -= 1.0;
+    }
+
+    return output;
+}
+
+} // namespace AudioCore

src/audio_core/algorithm/interpolate.h

@@ -0,0 +1,43 @@
+// Copyright 2018 yuzu Emulator Project
+// Licensed under GPLv2 or any later version
+// Refer to the license.txt file included.
+
+#pragma once
+
+#include <array>
+#include <vector>
+#include "audio_core/algorithm/filter.h"
+#include "common/common_types.h"
+
+namespace AudioCore {
+
+struct InterpolationState {
+    static constexpr size_t lanczos_taps = 4;
+    static constexpr size_t history_size = lanczos_taps * 2 - 1;
+
+    double current_ratio = 0.0;
+    CascadingFilter nyquist;
+    std::array<std::array<s16, 2>, history_size> history = {};
+    double position = 0;
+};
+
+/// Interpolates input signal to produce output signal.
+/// @param input The signal to interpolate.
+/// @param ratio Interpolation ratio.
+///              ratio > 1.0 results in fewer output samples.
+///              ratio < 1.0 results in more output samples.
+/// @returns Output signal.
+std::vector<s16> Interpolate(InterpolationState& state, std::vector<s16> input, double ratio);
+
+/// Interpolates input signal to produce output signal.
+/// @param input The signal to interpolate.
+/// @param input_rate The sample rate of input.
+/// @param output_rate The desired sample rate of the output.
+/// @returns Output signal.
+inline std::vector<s16> Interpolate(InterpolationState& state, std::vector<s16> input,
+                                    u32 input_rate, u32 output_rate) {
+    const double ratio = static_cast<double>(input_rate) / static_cast<double>(output_rate);
+    return Interpolate(state, std::move(input), ratio);
+}
+
+} // namespace AudioCore

src/audio_core/audio_out.cpp

@@ -0,0 +1,58 @@
+// Copyright 2018 yuzu Emulator Project
+// Licensed under GPLv2 or any later version
+// Refer to the license.txt file included.
+
+#include "audio_core/audio_out.h"
+#include "audio_core/sink.h"
+#include "audio_core/sink_details.h"
+#include "common/assert.h"
+#include "common/logging/log.h"
+#include "core/settings.h"
+
+namespace AudioCore {
+
+/// Returns the stream format from the specified number of channels
+static Stream::Format ChannelsToStreamFormat(u32 num_channels) {
+    switch (num_channels) {
+    case 1:
+        return Stream::Format::Mono16;
+    case 2:
+        return Stream::Format::Stereo16;
+    case 6:
+        return Stream::Format::Multi51Channel16;
+    }
+
+    LOG_CRITICAL(Audio, "Unimplemented num_channels={}", num_channels);
+    UNREACHABLE();
+    return {};
+}
+
+StreamPtr AudioOut::OpenStream(u32 sample_rate, u32 num_channels, std::string&& name,
+                               Stream::ReleaseCallback&& release_callback) {
+    if (!sink) {
+        const SinkDetails& sink_details = GetSinkDetails(Settings::values.sink_id);
+        sink = sink_details.factory(Settings::values.audio_device_id);
+    }
+
+    return std::make_shared<Stream>(
+        sample_rate, ChannelsToStreamFormat(num_channels), std::move(release_callback),
+        sink->AcquireSinkStream(sample_rate, num_channels, name), std::move(name));
+}
+
+std::vector<Buffer::Tag> AudioOut::GetTagsAndReleaseBuffers(StreamPtr stream, size_t max_count) {
+    return stream->GetTagsAndReleaseBuffers(max_count);
+}
+
+void AudioOut::StartStream(StreamPtr stream) {
+    stream->Play();
+}
+
+void AudioOut::StopStream(StreamPtr stream) {
+    stream->Stop();
+}
+
+bool AudioOut::QueueBuffer(StreamPtr stream, Buffer::Tag tag, std::vector<s16>&& data) {
+    return stream->QueueBuffer(std::make_shared<Buffer>(tag, std::move(data)));
+}
+
+} // namespace AudioCore

src/audio_core/audio_out.h

@@ -0,0 +1,43 @@
+// Copyright 2018 yuzu Emulator Project
+// Licensed under GPLv2 or any later version
+// Refer to the license.txt file included.
+
+#pragma once
+
+#include <memory>
+#include <string>
+#include <vector>
+
+#include "audio_core/buffer.h"
+#include "audio_core/sink.h"
+#include "audio_core/stream.h"
+#include "common/common_types.h"
+
+namespace AudioCore {
+
+/**
+ * Represents an audio playback interface, used to open and play audio streams
+ */
+class AudioOut {
+public:
+    /// Opens a new audio stream
+    StreamPtr OpenStream(u32 sample_rate, u32 num_channels, std::string&& name,
+                         Stream::ReleaseCallback&& release_callback);
+
+    /// Returns a vector of recently released buffers specified by tag for the specified stream
+    std::vector<Buffer::Tag> GetTagsAndReleaseBuffers(StreamPtr stream, size_t max_count);
+
+    /// Starts an audio stream for playback
+    void StartStream(StreamPtr stream);
+
+    /// Stops an audio stream that is currently playing
+    void StopStream(StreamPtr stream);
+
+    /// Queues a buffer into the specified audio stream, returns true on success
+    bool QueueBuffer(StreamPtr stream, Buffer::Tag tag, std::vector<s16>&& data);
+
+private:
+    SinkPtr sink;
+};
+
+} // namespace AudioCore

src/audio_core/audio_renderer.cpp

@@ -0,0 +1,249 @@
+// Copyright 2018 yuzu Emulator Project
+// Licensed under GPLv2 or any later version
+// Refer to the license.txt file included.
+
+#include "audio_core/algorithm/interpolate.h"
+#include "audio_core/audio_renderer.h"
+#include "common/assert.h"
+#include "common/logging/log.h"
+#include "core/memory.h"
+
+namespace AudioCore {
+
+constexpr u32 STREAM_SAMPLE_RATE{48000};
+constexpr u32 STREAM_NUM_CHANNELS{2};
+
+AudioRenderer::AudioRenderer(AudioRendererParameter params,
+                             Kernel::SharedPtr<Kernel::Event> buffer_event)
+    : worker_params{params}, buffer_event{buffer_event}, voices(params.voice_count) {
+
+    audio_core = std::make_unique<AudioCore::AudioOut>();
+    stream = audio_core->OpenStream(STREAM_SAMPLE_RATE, STREAM_NUM_CHANNELS, "AudioRenderer",
+                                    [=]() { buffer_event->Signal(); });
+    audio_core->StartStream(stream);
+
+    QueueMixedBuffer(0);
+    QueueMixedBuffer(1);
+    QueueMixedBuffer(2);
+}
+
+u32 AudioRenderer::GetSampleRate() const {
+    return worker_params.sample_rate;
+}
+
+u32 AudioRenderer::GetSampleCount() const {
+    return worker_params.sample_count;
+}
+
+u32 AudioRenderer::GetMixBufferCount() const {
+    return worker_params.mix_buffer_count;
+}
+
+std::vector<u8> AudioRenderer::UpdateAudioRenderer(const std::vector<u8>& input_params) {
+    // Copy UpdateDataHeader struct
+    UpdateDataHeader config{};
+    std::memcpy(&config, input_params.data(), sizeof(UpdateDataHeader));
+    u32 memory_pool_count = worker_params.effect_count + (worker_params.voice_count * 4);
+
+    // Copy MemoryPoolInfo structs
+    std::vector<MemoryPoolInfo> mem_pool_info(memory_pool_count);
+    std::memcpy(mem_pool_info.data(),
+                input_params.data() + sizeof(UpdateDataHeader) + config.behavior_size,
+                memory_pool_count * sizeof(MemoryPoolInfo));
+
+    // Copy VoiceInfo structs
+    size_t offset{sizeof(UpdateDataHeader) + config.behavior_size + config.memory_pools_size +
+                  config.voice_resource_size};
+    for (auto& voice : voices) {
+        std::memcpy(&voice.Info(), input_params.data() + offset, sizeof(VoiceInfo));
+        offset += sizeof(VoiceInfo);
+    }
+
+    // Update voices
+    for (auto& voice : voices) {
+        voice.UpdateState();
+        if (!voice.GetInfo().is_in_use) {
+            continue;
+        }
+        if (voice.GetInfo().is_new) {
+            voice.SetWaveIndex(voice.GetInfo().wave_buffer_head);
+        }
+    }
+
+    // Update memory pool state
+    std::vector<MemoryPoolEntry> memory_pool(memory_pool_count);
+    for (size_t index = 0; index < memory_pool.size(); ++index) {
+        if (mem_pool_info[index].pool_state == MemoryPoolStates::RequestAttach) {
+            memory_pool[index].state = MemoryPoolStates::Attached;
+        } else if (mem_pool_info[index].pool_state == MemoryPoolStates::RequestDetach) {
+            memory_pool[index].state = MemoryPoolStates::Detached;
+        }
+    }
+
+    // Release previous buffers and queue next ones for playback
+    ReleaseAndQueueBuffers();
+
+    // Copy output header
+    UpdateDataHeader response_data{worker_params};
+    std::vector<u8> output_params(response_data.total_size);
+    std::memcpy(output_params.data(), &response_data, sizeof(UpdateDataHeader));
+
+    // Copy output memory pool entries
+    std::memcpy(output_params.data() + sizeof(UpdateDataHeader), memory_pool.data(),
+                response_data.memory_pools_size);
+
+    // Copy output voice status
+    size_t voice_out_status_offset{sizeof(UpdateDataHeader) + response_data.memory_pools_size};
+    for (const auto& voice : voices) {
+        std::memcpy(output_params.data() + voice_out_status_offset, &voice.GetOutStatus(),
+                    sizeof(VoiceOutStatus));
+        voice_out_status_offset += sizeof(VoiceOutStatus);
+    }
+
+    return output_params;
+}
+
+void AudioRenderer::VoiceState::SetWaveIndex(size_t index) {
+    wave_index = index & 3;
+    is_refresh_pending = true;
+}
+
+std::vector<s16> AudioRenderer::VoiceState::DequeueSamples(size_t sample_count) {
+    if (!IsPlaying()) {
+        return {};
+    }
+
+    if (is_refresh_pending) {
+        RefreshBuffer();
+    }
+
+    const size_t max_size{samples.size() - offset};
+    const size_t dequeue_offset{offset};
+    size_t size{sample_count * STREAM_NUM_CHANNELS};
+    if (size > max_size) {
+        size = max_size;
+    }
+
+    out_status.played_sample_count += size / STREAM_NUM_CHANNELS;
+    offset += size;
+
+    const auto& wave_buffer{info.wave_buffer[wave_index]};
+    if (offset == samples.size()) {
+        offset = 0;
+
+        if (!wave_buffer.is_looping) {
+            SetWaveIndex(wave_index + 1);
+        }
+
+        out_status.wave_buffer_consumed++;
+
+        if (wave_buffer.end_of_stream) {
+            info.play_state = PlayState::Paused;
+        }
+    }
+
+    return {samples.begin() + dequeue_offset, samples.begin() + dequeue_offset + size};
+}
+
+void AudioRenderer::VoiceState::UpdateState() {
+    if (is_in_use && !info.is_in_use) {
+        // No longer in use, reset state
+        is_refresh_pending = true;
+        wave_index = 0;
+        offset = 0;
+        out_status = {};
+    }
+    is_in_use = info.is_in_use;
+}
+
+void AudioRenderer::VoiceState::RefreshBuffer() {
+    std::vector<s16> new_samples(info.wave_buffer[wave_index].buffer_sz / sizeof(s16));
+    Memory::ReadBlock(info.wave_buffer[wave_index].buffer_addr, new_samples.data(),
+                      info.wave_buffer[wave_index].buffer_sz);
+
+    switch (static_cast<Codec::PcmFormat>(info.sample_format)) {
+    case Codec::PcmFormat::Int16: {
+        // PCM16 is played as-is
+        break;
+    }
+    case Codec::PcmFormat::Adpcm: {
+        // Decode ADPCM to PCM16
+        Codec::ADPCM_Coeff coeffs;
+        Memory::ReadBlock(info.additional_params_addr, coeffs.data(), sizeof(Codec::ADPCM_Coeff));
+        new_samples = Codec::DecodeADPCM(reinterpret_cast<u8*>(new_samples.data()),
+                                         new_samples.size() * sizeof(s16), coeffs, adpcm_state);
+        break;
+    }
+    default:
+        LOG_CRITICAL(Audio, "Unimplemented sample_format={}", info.sample_format);
+        UNREACHABLE();
+        break;
+    }
+
+    switch (info.channel_count) {
+    case 1:
+        // 1 channel is upsampled to 2 channel
+        samples.resize(new_samples.size() * 2);
+        for (size_t index = 0; index < new_samples.size(); ++index) {
+            samples[index * 2] = new_samples[index];
+            samples[index * 2 + 1] = new_samples[index];
+        }
+        break;
+    case 2: {
+        // 2 channel is played as is
+        samples = std::move(new_samples);
+        break;
+    }
+    default:
+        LOG_CRITICAL(Audio, "Unimplemented channel_count={}", info.channel_count);
+        UNREACHABLE();
+        break;
+    }
+
+    samples = Interpolate(interp_state, std::move(samples), Info().sample_rate, STREAM_SAMPLE_RATE);
+
+    is_refresh_pending = false;
+}
+
+static constexpr s16 ClampToS16(s32 value) {
+    return static_cast<s16>(std::clamp(value, -32768, 32767));
+}
+
+void AudioRenderer::QueueMixedBuffer(Buffer::Tag tag) {
+    constexpr size_t BUFFER_SIZE{512};
+    std::vector<s16> buffer(BUFFER_SIZE * stream->GetNumChannels());
+
+    for (auto& voice : voices) {
+        if (!voice.IsPlaying()) {
+            continue;
+        }
+
+        size_t offset{};
+        s64 samples_remaining{BUFFER_SIZE};
+        while (samples_remaining > 0) {
+            const std::vector<s16> samples{voice.DequeueSamples(samples_remaining)};
+
+            if (samples.empty()) {
+                break;
+            }
+
+            samples_remaining -= samples.size() / stream->GetNumChannels();
+
+            for (const auto& sample : samples) {
+                const s32 buffer_sample{buffer[offset]};
+                buffer[offset++] =
+                    ClampToS16(buffer_sample + static_cast<s32>(sample * voice.GetInfo().volume));
+            }
+        }
+    }
+    audio_core->QueueBuffer(stream, tag, std::move(buffer));
+}
+
+void AudioRenderer::ReleaseAndQueueBuffers() {
+    const auto released_buffers{audio_core->GetTagsAndReleaseBuffers(stream, 2)};
+    for (const auto& tag : released_buffers) {
+        QueueMixedBuffer(tag);
+    }
+}
+
+} // namespace AudioCore

src/audio_core/audio_renderer.h

@@ -0,0 +1,211 @@
+// Copyright 2018 yuzu Emulator Project
+// Licensed under GPLv2 or any later version
+// Refer to the license.txt file included.
+
+#pragma once
+
+#include <array>
+#include <memory>
+#include <vector>
+
+#include "audio_core/algorithm/interpolate.h"
+#include "audio_core/audio_out.h"
+#include "audio_core/codec.h"
+#include "audio_core/stream.h"
+#include "common/common_types.h"
+#include "common/swap.h"
+#include "core/hle/kernel/event.h"
+
+namespace AudioCore {
+
+enum class PlayState : u8 {
+    Started = 0,
+    Stopped = 1,
+    Paused = 2,
+};
+
+struct AudioRendererParameter {
+    u32_le sample_rate;
+    u32_le sample_count;
+    u32_le mix_buffer_count;
+    u32_le unknown_c;
+    u32_le voice_count;
+    u32_le sink_count;
+    u32_le effect_count;
+    u32_le unknown_1c;
+    u8 unknown_20;
+    INSERT_PADDING_BYTES(3);
+    u32_le splitter_count;
+    u32_le unknown_2c;
+    INSERT_PADDING_WORDS(1);
+    u32_le revision;
+};
+static_assert(sizeof(AudioRendererParameter) == 52, "AudioRendererParameter is an invalid size");
+
+enum class MemoryPoolStates : u32 { // Should be LE
+    Invalid = 0x0,
+    Unknown = 0x1,
+    RequestDetach = 0x2,
+    Detached = 0x3,
+    RequestAttach = 0x4,
+    Attached = 0x5,
+    Released = 0x6,
+};
+
+struct MemoryPoolEntry {
+    MemoryPoolStates state;
+    u32_le unknown_4;
+    u32_le unknown_8;
+    u32_le unknown_c;
+};
+static_assert(sizeof(MemoryPoolEntry) == 0x10, "MemoryPoolEntry has wrong size");
+
+struct MemoryPoolInfo {
+    u64_le pool_address;
+    u64_le pool_size;
+    MemoryPoolStates pool_state;
+    INSERT_PADDING_WORDS(3); // Unknown
+};
+static_assert(sizeof(MemoryPoolInfo) == 0x20, "MemoryPoolInfo has wrong size");
+struct BiquadFilter {
+    u8 enable;
+    INSERT_PADDING_BYTES(1);
+    std::array<s16_le, 3> numerator;
+    std::array<s16_le, 2> denominator;
+};
+static_assert(sizeof(BiquadFilter) == 0xc, "BiquadFilter has wrong size");
+
+struct WaveBuffer {
+    u64_le buffer_addr;
+    u64_le buffer_sz;
+    s32_le start_sample_offset;
+    s32_le end_sample_offset;
+    u8 is_looping;
+    u8 end_of_stream;
+    u8 sent_to_server;
+    INSERT_PADDING_BYTES(5);
+    u64 context_addr;
+    u64 context_sz;
+    INSERT_PADDING_BYTES(8);
+};
+static_assert(sizeof(WaveBuffer) == 0x38, "WaveBuffer has wrong size");
+
+struct VoiceInfo {
+    u32_le id;
+    u32_le node_id;
+    u8 is_new;
+    u8 is_in_use;
+    PlayState play_state;
+    u8 sample_format;
+    u32_le sample_rate;
+    u32_le priority;
+    u32_le sorting_order;
+    u32_le channel_count;
+    float_le pitch;
+    float_le volume;
+    std::array<BiquadFilter, 2> biquad_filter;
+    u32_le wave_buffer_count;
+    u32_le wave_buffer_head;
+    INSERT_PADDING_WORDS(1);
+    u64_le additional_params_addr;
+    u64_le additional_params_sz;
+    u32_le mix_id;
+    u32_le splitter_info_id;
+    std::array<WaveBuffer, 4> wave_buffer;
+    std::array<u32_le, 6> voice_channel_resource_ids;
+    INSERT_PADDING_BYTES(24);
+};
+static_assert(sizeof(VoiceInfo) == 0x170, "VoiceInfo is wrong size");
+
+struct VoiceOutStatus {
+    u64_le played_sample_count;
+    u32_le wave_buffer_consumed;
+    u32_le voice_drops_count;
+};
+static_assert(sizeof(VoiceOutStatus) == 0x10, "VoiceOutStatus has wrong size");
+
+struct UpdateDataHeader {
+    UpdateDataHeader() {}
+
+    explicit UpdateDataHeader(const AudioRendererParameter& config) {
+        revision = Common::MakeMagic('R', 'E', 'V', '4'); // 5.1.0 Revision
+        behavior_size = 0xb0;
+        memory_pools_size = (config.effect_count + (config.voice_count * 4)) * 0x10;
+        voices_size = config.voice_count * 0x10;
+        voice_resource_size = 0x0;
+        effects_size = config.effect_count * 0x10;
+        mixes_size = 0x0;
+        sinks_size = config.sink_count * 0x20;
+        performance_manager_size = 0x10;
+        total_size = sizeof(UpdateDataHeader) + behavior_size + memory_pools_size + voices_size +
+                     effects_size + sinks_size + performance_manager_size;
+    }
+
+    u32_le revision;
+    u32_le behavior_size;
+    u32_le memory_pools_size;
+    u32_le voices_size;
+    u32_le voice_resource_size;
+    u32_le effects_size;
+    u32_le mixes_size;
+    u32_le sinks_size;
+    u32_le performance_manager_size;
+    INSERT_PADDING_WORDS(6);
+    u32_le total_size;
+};
+static_assert(sizeof(UpdateDataHeader) == 0x40, "UpdateDataHeader has wrong size");
+
+class AudioRenderer {
+public:
+    AudioRenderer(AudioRendererParameter params, Kernel::SharedPtr<Kernel::Event> buffer_event);
+    std::vector<u8> UpdateAudioRenderer(const std::vector<u8>& input_params);
+    void QueueMixedBuffer(Buffer::Tag tag);
+    void ReleaseAndQueueBuffers();
+    u32 GetSampleRate() const;
+    u32 GetSampleCount() const;
+    u32 GetMixBufferCount() const;
+
+private:
+    class VoiceState {
+    public:
+        bool IsPlaying() const {
+            return is_in_use && info.play_state == PlayState::Started;
+        }
+
+        const VoiceOutStatus& GetOutStatus() const {
+            return out_status;
+        }
+
+        const VoiceInfo& GetInfo() const {
+            return info;
+        }
+
+        VoiceInfo& Info() {
+            return info;
+        }
+
+        void SetWaveIndex(size_t index);
+        std::vector<s16> DequeueSamples(size_t sample_count);
+        void UpdateState();
+        void RefreshBuffer();
+
+    private:
+        bool is_in_use{};
+        bool is_refresh_pending{};
+        size_t wave_index{};
+        size_t offset{};
+        Codec::ADPCMState adpcm_state{};
+        InterpolationState interp_state{};
+        std::vector<s16> samples;
+        VoiceOutStatus out_status{};
+        VoiceInfo info{};
+    };
+
+    AudioRendererParameter worker_params;
+    Kernel::SharedPtr<Kernel::Event> buffer_event;
+    std::vector<VoiceState> voices;
+    std::unique_ptr<AudioCore::AudioOut> audio_core;
+    AudioCore::StreamPtr stream;
+};
+
+} // namespace AudioCore

src/audio_core/buffer.h

@@ -0,0 +1,45 @@
+// Copyright 2018 yuzu Emulator Project
+// Licensed under GPLv2 or any later version
+// Refer to the license.txt file included.
+
+#pragma once
+
+#include <memory>
+#include <vector>
+
+#include "common/common_types.h"
+
+namespace AudioCore {
+
+/**
+ * Represents a buffer of audio samples to be played in an audio stream
+ */
+class Buffer {
+public:
+    using Tag = u64;
+
+    Buffer(Tag tag, std::vector<s16>&& samples) : tag{tag}, samples{std::move(samples)} {}
+
+    /// Returns the raw audio data for the buffer
+    std::vector<s16>& Samples() {
+        return samples;
+    }
+
+    /// Returns the raw audio data for the buffer
+    const std::vector<s16>& GetSamples() const {
+        return samples;
+    }
+
+    /// Returns the buffer tag, this is provided by the game to the audout service
+    Tag GetTag() const {
+        return tag;
+    }
+
+private:
+    Tag tag;
+    std::vector<s16> samples;
+};
+
+using BufferPtr = std::shared_ptr<Buffer>;
+
+} // namespace AudioCore

src/audio_core/codec.cpp

@@ -0,0 +1,77 @@
+// Copyright 2018 yuzu Emulator Project
+// Licensed under GPLv2 or any later version
+// Refer to the license.txt file included.
+
+#include <algorithm>
+
+#include "audio_core/codec.h"
+
+namespace AudioCore::Codec {
+
+std::vector<s16> DecodeADPCM(const u8* const data, size_t size, const ADPCM_Coeff& coeff,
+                             ADPCMState& state) {
+    // GC-ADPCM with scale factor and variable coefficients.
+    // Frames are 8 bytes long containing 14 samples each.
+    // Samples are 4 bits (one nibble) long.
+
+    constexpr size_t FRAME_LEN = 8;
+    constexpr size_t SAMPLES_PER_FRAME = 14;
+    constexpr std::array<int, 16> SIGNED_NIBBLES = {
+        {0, 1, 2, 3, 4, 5, 6, 7, -8, -7, -6, -5, -4, -3, -2, -1}};
+
+    const size_t sample_count = (size / FRAME_LEN) * SAMPLES_PER_FRAME;
+    const size_t ret_size =
+        sample_count % 2 == 0 ? sample_count : sample_count + 1; // Ensure multiple of two.
+    std::vector<s16> ret(ret_size);
+
+    int yn1 = state.yn1, yn2 = state.yn2;
+
+    const size_t NUM_FRAMES =
+        (sample_count + (SAMPLES_PER_FRAME - 1)) / SAMPLES_PER_FRAME; // Round up.
+    for (size_t framei = 0; framei < NUM_FRAMES; framei++) {
+        const int frame_header = data[framei * FRAME_LEN];
+        const int scale = 1 << (frame_header & 0xF);
+        const int idx = (frame_header >> 4) & 0x7;
+
+        // Coefficients are fixed point with 11 bits fractional part.
+        const int coef1 = coeff[idx * 2 + 0];
+        const int coef2 = coeff[idx * 2 + 1];
+
+        // Decodes an audio sample. One nibble produces one sample.
+        const auto decode_sample = [&](const int nibble) -> s16 {
+            const int xn = nibble * scale;
+            // We first transform everything into 11 bit fixed point, perform the second order
+            // digital filter, then transform back.
+            // 0x400 == 0.5 in 11 bit fixed point.
+            // Filter: y[n] = x[n] + 0.5 + c1 * y[n-1] + c2 * y[n-2]
+            int val = ((xn << 11) + 0x400 + coef1 * yn1 + coef2 * yn2) >> 11;
+            // Clamp to output range.
+            val = std::clamp<s32>(val, -32768, 32767);
+            // Advance output feedback.
+            yn2 = yn1;
+            yn1 = val;
+            return static_cast<s16>(val);
+        };
+
+        size_t outputi = framei * SAMPLES_PER_FRAME;
+        size_t datai = framei * FRAME_LEN + 1;
+        for (size_t i = 0; i < SAMPLES_PER_FRAME && outputi < sample_count; i += 2) {
+            const s16 sample1 = decode_sample(SIGNED_NIBBLES[data[datai] >> 4]);
+            ret[outputi] = sample1;
+            outputi++;
+
+            const s16 sample2 = decode_sample(SIGNED_NIBBLES[data[datai] & 0xF]);
+            ret[outputi] = sample2;
+            outputi++;
+
+            datai++;
+        }
+    }
+
+    state.yn1 = yn1;
+    state.yn2 = yn2;
+
+    return ret;
+}
+
+} // namespace AudioCore::Codec

src/audio_core/codec.h

@@ -0,0 +1,44 @@
+// Copyright 2018 yuzu Emulator Project
+// Licensed under GPLv2 or any later version
+// Refer to the license.txt file included.
+
+#pragma once
+
+#include <array>
+#include <vector>
+
+#include "common/common_types.h"
+
+namespace AudioCore::Codec {
+
+enum class PcmFormat : u32 {
+    Invalid = 0,
+    Int8 = 1,
+    Int16 = 2,
+    Int24 = 3,
+    Int32 = 4,
+    PcmFloat = 5,
+    Adpcm = 6,
+};
+
+/// See: Codec::DecodeADPCM
+struct ADPCMState {
+    // Two historical samples from previous processed buffer,
+    // required for ADPCM decoding
+    s16 yn1; ///< y[n-1]
+    s16 yn2; ///< y[n-2]
+};
+
+using ADPCM_Coeff = std::array<s16, 16>;
+
+/**
+ * @param data Pointer to buffer that contains ADPCM data to decode
+ * @param size Size of buffer in bytes
+ * @param coeff ADPCM coefficients
+ * @param state ADPCM state, this is updated with new state
+ * @return Decoded stereo signed PCM16 data, sample_count in length
+ */
+std::vector<s16> DecodeADPCM(const u8* const data, size_t size, const ADPCM_Coeff& coeff,
+                             ADPCMState& state);
+
+}; // namespace AudioCore::Codec

src/audio_core/cubeb_sink.cpp

@@ -0,0 +1,206 @@
+// Copyright 2018 yuzu Emulator Project
+// Licensed under GPLv2 or any later version
+// Refer to the license.txt file included.
+
+#include <algorithm>
+#include <cstring>
+#include <mutex>
+
+#include "audio_core/cubeb_sink.h"
+#include "audio_core/stream.h"
+#include "common/logging/log.h"
+
+namespace AudioCore {
+
+class SinkStreamImpl final : public SinkStream {
+public:
+    SinkStreamImpl(cubeb* ctx, u32 sample_rate, u32 num_channels_, cubeb_devid output_device,
+                   const std::string& name)
+        : ctx{ctx}, num_channels{num_channels_} {
+
+        if (num_channels == 6) {
+            // 6-channel audio does not seem to work with cubeb + SDL, so we downsample this to 2
+            // channel for now
+            is_6_channel = true;
+            num_channels = 2;
+        }
+
+        cubeb_stream_params params{};
+        params.rate = sample_rate;
+        params.channels = num_channels;
+        params.format = CUBEB_SAMPLE_S16NE;
+        params.layout = num_channels == 1 ? CUBEB_LAYOUT_MONO : CUBEB_LAYOUT_STEREO;
+
+        u32 minimum_latency{};
+        if (cubeb_get_min_latency(ctx, &params, &minimum_latency) != CUBEB_OK) {
+            LOG_CRITICAL(Audio_Sink, "Error getting minimum latency");
+        }
+
+        if (cubeb_stream_init(ctx, &stream_backend, name.c_str(), nullptr, nullptr, output_device,
+                              &params, std::max(512u, minimum_latency),
+                              &SinkStreamImpl::DataCallback, &SinkStreamImpl::StateCallback,
+                              this) != CUBEB_OK) {
+            LOG_CRITICAL(Audio_Sink, "Error initializing cubeb stream");
+            return;
+        }
+
+        if (cubeb_stream_start(stream_backend) != CUBEB_OK) {
+            LOG_CRITICAL(Audio_Sink, "Error starting cubeb stream");
+            return;
+        }
+    }
+
+    ~SinkStreamImpl() {
+        if (!ctx) {
+            return;
+        }
+
+        if (cubeb_stream_stop(stream_backend) != CUBEB_OK) {
+            LOG_CRITICAL(Audio_Sink, "Error stopping cubeb stream");
+        }
+
+        cubeb_stream_destroy(stream_backend);
+    }
+
+    void EnqueueSamples(u32 num_channels, const std::vector<s16>& samples) override {
+        if (!ctx) {
+            return;
+        }
+
+        std::lock_guard lock{queue_mutex};
+
+        queue.reserve(queue.size() + samples.size() * GetNumChannels());
+
+        if (is_6_channel) {
+            // Downsample 6 channels to 2
+            const size_t sample_count_copy_size = samples.size() * 2;
+            queue.reserve(sample_count_copy_size);
+            for (size_t i = 0; i < samples.size(); i += num_channels) {
+                queue.push_back(samples[i]);
+                queue.push_back(samples[i + 1]);
+            }
+        } else {
+            // Copy as-is
+            std::copy(samples.begin(), samples.end(), std::back_inserter(queue));
+        }
+    }
+
+    u32 GetNumChannels() const {
+        return num_channels;
+    }
+
+private:
+    std::vector<std::string> device_list;
+
+    cubeb* ctx{};
+    cubeb_stream* stream_backend{};
+    u32 num_channels{};
+    bool is_6_channel{};
+
+    std::mutex queue_mutex;
+    std::vector<s16> queue;
+
+    static long DataCallback(cubeb_stream* stream, void* user_data, const void* input_buffer,
+                             void* output_buffer, long num_frames);
+    static void StateCallback(cubeb_stream* stream, void* user_data, cubeb_state state);
+};
+
+CubebSink::CubebSink(std::string target_device_name) {
+    if (cubeb_init(&ctx, "yuzu", nullptr) != CUBEB_OK) {
+        LOG_CRITICAL(Audio_Sink, "cubeb_init failed");
+        return;
+    }
+
+    if (target_device_name != auto_device_name && !target_device_name.empty()) {
+        cubeb_device_collection collection;
+        if (cubeb_enumerate_devices(ctx, CUBEB_DEVICE_TYPE_OUTPUT, &collection) != CUBEB_OK) {
+            LOG_WARNING(Audio_Sink, "Audio output device enumeration not supported");
+        } else {
+            const auto collection_end{collection.device + collection.count};
+            const auto device{std::find_if(collection.device, collection_end,
+                                           [&](const cubeb_device_info& device) {
+                                               return target_device_name == device.friendly_name;
+                                           })};
+            if (device != collection_end) {
+                output_device = device->devid;
+            }
+            cubeb_device_collection_destroy(ctx, &collection);
+        }
+    }
+}
+
+CubebSink::~CubebSink() {
+    if (!ctx) {
+        return;
+    }
+
+    for (auto& sink_stream : sink_streams) {
+        sink_stream.reset();
+    }
+
+    cubeb_destroy(ctx);
+}
+
+SinkStream& CubebSink::AcquireSinkStream(u32 sample_rate, u32 num_channels,
+                                         const std::string& name) {
+    sink_streams.push_back(
+        std::make_unique<SinkStreamImpl>(ctx, sample_rate, num_channels, output_device, name));
+    return *sink_streams.back();
+}
+
+long SinkStreamImpl::DataCallback(cubeb_stream* stream, void* user_data, const void* input_buffer,
+                                  void* output_buffer, long num_frames) {
+    SinkStreamImpl* impl = static_cast<SinkStreamImpl*>(user_data);
+    u8* buffer = reinterpret_cast<u8*>(output_buffer);
+
+    if (!impl) {
+        return {};
+    }
+
+    std::lock_guard lock{impl->queue_mutex};
+
+    const size_t frames_to_write{
+        std::min(impl->queue.size() / impl->GetNumChannels(), static_cast<size_t>(num_frames))};
+
+    memcpy(buffer, impl->queue.data(), frames_to_write * sizeof(s16) * impl->GetNumChannels());
+    impl->queue.erase(impl->queue.begin(),
+                      impl->queue.begin() + frames_to_write * impl->GetNumChannels());
+
+    if (frames_to_write < num_frames) {
+        // Fill the rest of the frames with silence
+        memset(buffer + frames_to_write * sizeof(s16) * impl->GetNumChannels(), 0,
+               (num_frames - frames_to_write) * sizeof(s16) * impl->GetNumChannels());
+    }
+
+    return num_frames;
+}
+
+void SinkStreamImpl::StateCallback(cubeb_stream* stream, void* user_data, cubeb_state state) {}
+
+std::vector<std::string> ListCubebSinkDevices() {
+    std::vector<std::string> device_list;
+    cubeb* ctx;
+
+    if (cubeb_init(&ctx, "Citra Device Enumerator", nullptr) != CUBEB_OK) {
+        LOG_CRITICAL(Audio_Sink, "cubeb_init failed");
+        return {};
+    }
+
+    cubeb_device_collection collection;
+    if (cubeb_enumerate_devices(ctx, CUBEB_DEVICE_TYPE_OUTPUT, &collection) != CUBEB_OK) {
+        LOG_WARNING(Audio_Sink, "Audio output device enumeration not supported");
+    } else {
+        for (size_t i = 0; i < collection.count; i++) {
+            const cubeb_device_info& device = collection.device[i];
+            if (device.friendly_name) {
+                device_list.emplace_back(device.friendly_name);
+            }
+        }
+        cubeb_device_collection_destroy(ctx, &collection);
+    }
+
+    cubeb_destroy(ctx);
+    return device_list;
+}
+
+} // namespace AudioCore

src/audio_core/cubeb_sink.h

@@ -0,0 +1,32 @@
+// Copyright 2018 yuzu Emulator Project
+// Licensed under GPLv2 or any later version
+// Refer to the license.txt file included.
+
+#pragma once
+
+#include <string>
+#include <vector>
+
+#include <cubeb/cubeb.h>
+
+#include "audio_core/sink.h"
+
+namespace AudioCore {
+
+class CubebSink final : public Sink {
+public:
+    explicit CubebSink(std::string device_id);
+    ~CubebSink() override;
+
+    SinkStream& AcquireSinkStream(u32 sample_rate, u32 num_channels,
+                                  const std::string& name) override;
+
+private:
+    cubeb* ctx{};
+    cubeb_devid output_device{};
+    std::vector<SinkStreamPtr> sink_streams;
+};
+
+std::vector<std::string> ListCubebSinkDevices();
+
+} // namespace AudioCore

src/audio_core/null_sink.h

@@ -0,0 +1,27 @@
+// Copyright 2018 yuzu Emulator Project
+// Licensed under GPLv2 or any later version
+// Refer to the license.txt file included.
+
+#pragma once
+
+#include "audio_core/sink.h"
+
+namespace AudioCore {
+
+class NullSink final : public Sink {
+public:
+    explicit NullSink(std::string){};
+    ~NullSink() override = default;
+
+    SinkStream& AcquireSinkStream(u32 /*sample_rate*/, u32 /*num_channels*/,
+                                  const std::string& /*name*/) override {
+        return null_sink_stream;
+    }
+
+private:
+    struct NullSinkStreamImpl final : SinkStream {
+        void EnqueueSamples(u32 /*num_channels*/, const std::vector<s16>& /*samples*/) override {}
+    } null_sink_stream;
+};
+
+} // namespace AudioCore

src/audio_core/sink.h

@@ -0,0 +1,31 @@
+// Copyright 2018 yuzu Emulator Project
+// Licensed under GPLv2 or any later version
+// Refer to the license.txt file included.
+
+#pragma once
+
+#include <memory>
+#include <string>
+
+#include "audio_core/sink_stream.h"
+#include "common/common_types.h"
+
+namespace AudioCore {
+
+constexpr char auto_device_name[] = "auto";
+
+/**
+ * This class is an interface for an audio sink. An audio sink accepts samples in stereo signed
+ * PCM16 format to be output. Sinks *do not* handle resampling and expect the correct sample rate.
+ * They are dumb outputs.
+ */
+class Sink {
+public:
+    virtual ~Sink() = default;
+    virtual SinkStream& AcquireSinkStream(u32 sample_rate, u32 num_channels,
+                                          const std::string& name) = 0;
+};
+
+using SinkPtr = std::unique_ptr<Sink>;
+
+} // namespace AudioCore

src/audio_core/sink_details.cpp

@@ -0,0 +1,44 @@
+// Copyright 2018 yuzu Emulator Project
+// Licensed under GPLv2 or any later version
+// Refer to the license.txt file included.
+
+#include <algorithm>
+#include <memory>
+#include <string>
+#include <vector>
+#include "audio_core/null_sink.h"
+#include "audio_core/sink_details.h"
+#ifdef HAVE_CUBEB
+#include "audio_core/cubeb_sink.h"
+#endif
+#include "common/logging/log.h"
+
+namespace AudioCore {
+
+// g_sink_details is ordered in terms of desirability, with the best choice at the top.
+const std::vector<SinkDetails> g_sink_details = {
+#ifdef HAVE_CUBEB
+    SinkDetails{"cubeb", &std::make_unique<CubebSink, std::string>, &ListCubebSinkDevices},
+#endif
+    SinkDetails{"null", &std::make_unique<NullSink, std::string>,
+                [] { return std::vector<std::string>{"null"}; }},
+};
+
+const SinkDetails& GetSinkDetails(std::string sink_id) {
+    auto iter =
+        std::find_if(g_sink_details.begin(), g_sink_details.end(),
+                     [sink_id](const auto& sink_detail) { return sink_detail.id == sink_id; });
+
+    if (sink_id == "auto" || iter == g_sink_details.end()) {
+        if (sink_id != "auto") {
+            LOG_ERROR(Audio, "AudioCore::SelectSink given invalid sink_id {}", sink_id);
+        }
+        // Auto-select.
+        // g_sink_details is ordered in terms of desirability, with the best choice at the front.
+        iter = g_sink_details.begin();
+    }
+
+    return *iter;
+}
+
+} // namespace AudioCore

src/audio_core/sink_details.h

@@ -0,0 +1,35 @@
+// Copyright 2018 yuzu Emulator Project
+// Licensed under GPLv2 or any later version
+// Refer to the license.txt file included.
+
+#pragma once
+
+#include <functional>
+#include <memory>
+#include <utility>
+#include <vector>
+
+namespace AudioCore {
+
+class Sink;
+
+struct SinkDetails {
+    using FactoryFn = std::function<std::unique_ptr<Sink>(std::string)>;
+    using ListDevicesFn = std::function<std::vector<std::string>()>;
+
+    SinkDetails(const char* id_, FactoryFn factory_, ListDevicesFn list_devices_)
+        : id(id_), factory(std::move(factory_)), list_devices(std::move(list_devices_)) {}
+
+    /// Name for this sink.
+    const char* id;
+    /// A method to call to construct an instance of this type of sink.
+    FactoryFn factory;
+    /// A method to call to list available devices.
+    ListDevicesFn list_devices;
+};
+
+extern const std::vector<SinkDetails> g_sink_details;
+
+const SinkDetails& GetSinkDetails(std::string sink_id);
+
+} // namespace AudioCore

src/audio_core/sink_stream.h

@@ -0,0 +1,32 @@
+// Copyright 2018 yuzu Emulator Project
+// Licensed under GPLv2 or any later version
+// Refer to the license.txt file included.
+
+#pragma once
+
+#include <memory>
+#include <vector>
+
+#include "common/common_types.h"
+
+namespace AudioCore {
+
+/**
+ * Accepts samples in stereo signed PCM16 format to be output. Sinks *do not* handle resampling and
+ * expect the correct sample rate. They are dumb outputs.
+ */
+class SinkStream {
+public:
+    virtual ~SinkStream() = default;
+
+    /**
+     * Feed stereo samples to sink.
+     * @param num_channels Number of channels used.
+     * @param samples Samples in interleaved stereo PCM16 format.
+     */
+    virtual void EnqueueSamples(u32 num_channels, const std::vector<s16>& samples) = 0;
+};
+
+using SinkStreamPtr = std::unique_ptr<SinkStream>;
+
+} // namespace AudioCore

src/audio_core/stream.cpp

@@ -0,0 +1,127 @@
+// Copyright 2018 yuzu Emulator Project
+// Licensed under GPLv2 or any later version
+// Refer to the license.txt file included.
+
+#include <algorithm>
+#include <cmath>
+
+#include "audio_core/sink.h"
+#include "audio_core/sink_details.h"
+#include "audio_core/stream.h"
+#include "common/assert.h"
+#include "common/logging/log.h"
+#include "core/core_timing.h"
+#include "core/core_timing_util.h"
+#include "core/settings.h"
+
+namespace AudioCore {
+
+constexpr size_t MaxAudioBufferCount{32};
+
+u32 Stream::GetNumChannels() const {
+    switch (format) {
+    case Format::Mono16:
+        return 1;
+    case Format::Stereo16:
+        return 2;
+    case Format::Multi51Channel16:
+        return 6;
+    }
+    LOG_CRITICAL(Audio, "Unimplemented format={}", static_cast<u32>(format));
+    UNREACHABLE();
+    return {};
+}
+
+Stream::Stream(u32 sample_rate, Format format, ReleaseCallback&& release_callback,
+               SinkStream& sink_stream, std::string&& name_)
+    : sample_rate{sample_rate}, format{format}, release_callback{std::move(release_callback)},
+      sink_stream{sink_stream}, name{std::move(name_)} {
+
+    release_event = CoreTiming::RegisterEvent(
+        name, [this](u64 userdata, int cycles_late) { ReleaseActiveBuffer(); });
+}
+
+void Stream::Play() {
+    state = State::Playing;
+    PlayNextBuffer();
+}
+
+void Stream::Stop() {
+    ASSERT_MSG(false, "Unimplemented");
+}
+
+s64 Stream::GetBufferReleaseCycles(const Buffer& buffer) const {
+    const size_t num_samples{buffer.GetSamples().size() / GetNumChannels()};
+    return CoreTiming::usToCycles((static_cast<u64>(num_samples) * 1000000) / sample_rate);
+}
+
+static void VolumeAdjustSamples(std::vector<s16>& samples) {
+    const float volume{std::clamp(Settings::values.volume, 0.0f, 1.0f)};
+
+    if (volume == 1.0f) {
+        return;
+    }
+
+    // Implementation of a volume slider with a dynamic range of 60 dB
+    const float volume_scale_factor{std::exp(6.90775f * volume) * 0.001f};
+    for (auto& sample : samples) {
+        sample = static_cast<s16>(sample * volume_scale_factor);
+    }
+}
+
+void Stream::PlayNextBuffer() {
+    if (!IsPlaying()) {
+        // Ensure we are in playing state before playing the next buffer
+        return;
+    }
+
+    if (active_buffer) {
+        // Do not queue a new buffer if we are already playing a buffer
+        return;
+    }
+
+    if (queued_buffers.empty()) {
+        // No queued buffers - we are effectively paused
+        return;
+    }
+
+    active_buffer = queued_buffers.front();
+    queued_buffers.pop();
+
+    VolumeAdjustSamples(active_buffer->Samples());
+    sink_stream.EnqueueSamples(GetNumChannels(), active_buffer->GetSamples());
+
+    CoreTiming::ScheduleEventThreadsafe(GetBufferReleaseCycles(*active_buffer), release_event, {});
+}
+
+void Stream::ReleaseActiveBuffer() {
+    ASSERT(active_buffer);
+    released_buffers.push(std::move(active_buffer));
+    release_callback();
+    PlayNextBuffer();
+}
+
+bool Stream::QueueBuffer(BufferPtr&& buffer) {
+    if (queued_buffers.size() < MaxAudioBufferCount) {
+        queued_buffers.push(std::move(buffer));
+        PlayNextBuffer();
+        return true;
+    }
+    return false;
+}
+
+bool Stream::ContainsBuffer(Buffer::Tag tag) const {
+    ASSERT_MSG(false, "Unimplemented");
+    return {};
+}
+
+std::vector<Buffer::Tag> Stream::GetTagsAndReleaseBuffers(size_t max_count) {
+    std::vector<Buffer::Tag> tags;
+    for (size_t count = 0; count < max_count && !released_buffers.empty(); ++count) {
+        tags.push_back(released_buffers.front()->GetTag());
+        released_buffers.pop();
+    }
+    return tags;
+}
+
+} // namespace AudioCore

src/audio_core/stream.h

@@ -0,0 +1,102 @@
+// Copyright 2018 yuzu Emulator Project
+// Licensed under GPLv2 or any later version
+// Refer to the license.txt file included.
+
+#pragma once
+
+#include <functional>
+#include <memory>
+#include <string>
+#include <vector>
+#include <queue>
+
+#include "audio_core/buffer.h"
+#include "audio_core/sink_stream.h"
+#include "common/assert.h"
+#include "common/common_types.h"
+#include "core/core_timing.h"
+
+namespace AudioCore {
+
+/**
+ * Represents an audio stream, which is a sequence of queued buffers, to be outputed by AudioOut
+ */
+class Stream {
+public:
+    /// Audio format of the stream
+    enum class Format {
+        Mono16,
+        Stereo16,
+        Multi51Channel16,
+    };
+
+    /// Callback function type, used to change guest state on a buffer being released
+    using ReleaseCallback = std::function<void()>;
+
+    Stream(u32 sample_rate, Format format, ReleaseCallback&& release_callback,
+           SinkStream& sink_stream, std::string&& name_);
+
+    /// Plays the audio stream
+    void Play();
+
+    /// Stops the audio stream
+    void Stop();
+
+    /// Queues a buffer into the audio stream, returns true on success
+    bool QueueBuffer(BufferPtr&& buffer);
+
+    /// Returns true if the audio stream contains a buffer with the specified tag
+    bool ContainsBuffer(Buffer::Tag tag) const;
+
+    /// Returns a vector of recently released buffers specified by tag
+    std::vector<Buffer::Tag> GetTagsAndReleaseBuffers(size_t max_count);
+
+    /// Returns true if the stream is currently playing
+    bool IsPlaying() const {
+        return state == State::Playing;
+    }
+
+    /// Returns the number of queued buffers
+    size_t GetQueueSize() const {
+        return queued_buffers.size();
+    }
+
+    /// Gets the sample rate
+    u32 GetSampleRate() const {
+        return sample_rate;
+    }
+
+    /// Gets the number of channels
+    u32 GetNumChannels() const;
+
+private:
+    /// Current state of the stream
+    enum class State {
+        Stopped,
+        Playing,
+    };
+
+    /// Plays the next queued buffer in the audio stream, starting playback if necessary
+    void PlayNextBuffer();
+
+    /// Releases the actively playing buffer, signalling that it has been completed
+    void ReleaseActiveBuffer();
+
+    /// Gets the number of core cycles when the specified buffer will be released
+    s64 GetBufferReleaseCycles(const Buffer& buffer) const;
+
+    u32 sample_rate;                        ///< Sample rate of the stream
+    Format format;                          ///< Format of the stream
+    ReleaseCallback release_callback;       ///< Buffer release callback for the stream
+    State state{State::Stopped};            ///< Playback state of the stream
+    CoreTiming::EventType* release_event{}; ///< Core timing release event for the stream
+    BufferPtr active_buffer;                ///< Actively playing buffer in the stream
+    std::queue<BufferPtr> queued_buffers;   ///< Buffers queued to be played in the stream
+    std::queue<BufferPtr> released_buffers; ///< Buffers recently released from the stream
+    SinkStream& sink_stream;                ///< Output sink for the stream
+    std::string name;                       ///< Name of the stream, must be unique
+};
+
+using StreamPtr = std::shared_ptr<Stream>;
+
+} // namespace AudioCore

src/common/CMakeLists.txt

@@ -29,8 +29,6 @@ add_library(common STATIC
     assert.h
     bit_field.h
     bit_set.h
-    break_points.cpp
-    break_points.h
     cityhash.cpp
     cityhash.h
     color.h
@@ -40,6 +38,8 @@ add_library(common STATIC
     file_util.cpp
     file_util.h
     hash.h
+    hex_util.cpp
+    hex_util.h
     logging/backend.cpp
     logging/backend.h
     logging/filter.cpp
@@ -63,7 +63,6 @@ add_library(common STATIC
     string_util.cpp
     string_util.h
     swap.h
-    synchronized_wrapper.h
     telemetry.cpp
     telemetry.h
     thread.cpp

src/common/alignment.h

@@ -9,13 +9,13 @@ namespace Common {
 
 template <typename T>
 constexpr T AlignUp(T value, size_t size) {
-    static_assert(std::is_unsigned<T>::value, "T must be an unsigned value.");
+    static_assert(std::is_unsigned_v<T>, "T must be an unsigned value.");
     return static_cast<T>(value + (size - value % size) % size);
 }
 
 template <typename T>
 constexpr T AlignDown(T value, size_t size) {
-    static_assert(std::is_unsigned<T>::value, "T must be an unsigned value.");
+    static_assert(std::is_unsigned_v<T>, "T must be an unsigned value.");
     return static_cast<T>(value - value % size);
 }
 

src/common/bit_field.h

@@ -178,8 +178,7 @@ public:
         return ExtractValue(storage);
     }
 
-    // TODO: we may want to change this to explicit operator bool() if it's bug-free in VS2015
-    constexpr FORCE_INLINE bool ToBool() const {
+    constexpr explicit operator bool() const {
         return Value() != 0;
     }
 

src/common/bit_set.h

@@ -96,7 +96,7 @@ static inline int LeastSignificantSetBit(u64 val) {
 
 template <typename IntTy>
 class BitSet {
-    static_assert(!std::is_signed<IntTy>::value, "BitSet should not be used with signed types");
+    static_assert(!std::is_signed_v<IntTy>, "BitSet should not be used with signed types");
 
 public:
     // A reference to a particular bit, returned from operator[].

src/common/break_points.cpp

@@ -1,90 +0,0 @@
-// Copyright 2013 Dolphin Emulator Project / 2014 Citra Emulator Project
-// Licensed under GPLv2 or any later version
-// Refer to the license.txt file included.
-
-#include <algorithm>
-#include <sstream>
-#include "common/break_points.h"
-
-bool BreakPoints::IsAddressBreakPoint(u32 iAddress) const {
-    auto cond = [&iAddress](const TBreakPoint& bp) { return bp.iAddress == iAddress; };
-    auto it = std::find_if(m_BreakPoints.begin(), m_BreakPoints.end(), cond);
-    return it != m_BreakPoints.end();
-}
-
-bool BreakPoints::IsTempBreakPoint(u32 iAddress) const {
-    auto cond = [&iAddress](const TBreakPoint& bp) {
-        return bp.iAddress == iAddress && bp.bTemporary;
-    };
-    auto it = std::find_if(m_BreakPoints.begin(), m_BreakPoints.end(), cond);
-    return it != m_BreakPoints.end();
-}
-
-BreakPoints::TBreakPointsStr BreakPoints::GetStrings() const {
-    TBreakPointsStr bps;
-    for (auto breakpoint : m_BreakPoints) {
-        if (!breakpoint.bTemporary) {
-            std::stringstream bp;
-            bp << std::hex << breakpoint.iAddress << " " << (breakpoint.bOn ? "n" : "");
-            bps.push_back(bp.str());
-        }
-    }
-
-    return bps;
-}
-
-void BreakPoints::AddFromStrings(const TBreakPointsStr& bps) {
-    for (auto bps_item : bps) {
-        TBreakPoint bp;
-        std::stringstream bpstr;
-        bpstr << std::hex << bps_item;
-        bpstr >> bp.iAddress;
-        bp.bOn = bps_item.find("n") != bps_item.npos;
-        bp.bTemporary = false;
-        Add(bp);
-    }
-}
-
-void BreakPoints::Add(const TBreakPoint& bp) {
-    if (!IsAddressBreakPoint(bp.iAddress)) {
-        m_BreakPoints.push_back(bp);
-        // if (jit)
-        //    jit->GetBlockCache()->InvalidateICache(bp.iAddress, 4);
-    }
-}
-
-void BreakPoints::Add(u32 em_address, bool temp) {
-    if (!IsAddressBreakPoint(em_address)) // only add new addresses
-    {
-        TBreakPoint pt; // breakpoint settings
-        pt.bOn = true;
-        pt.bTemporary = temp;
-        pt.iAddress = em_address;
-
-        m_BreakPoints.push_back(pt);
-
-        // if (jit)
-        //    jit->GetBlockCache()->InvalidateICache(em_address, 4);
-    }
-}
-
-void BreakPoints::Remove(u32 em_address) {
-    auto cond = [&em_address](const TBreakPoint& bp) { return bp.iAddress == em_address; };
-    auto it = std::find_if(m_BreakPoints.begin(), m_BreakPoints.end(), cond);
-    if (it != m_BreakPoints.end())
-        m_BreakPoints.erase(it);
-}
-
-void BreakPoints::Clear() {
-    // if (jit)
-    //{
-    //    std::for_each(m_BreakPoints.begin(), m_BreakPoints.end(),
-    //        [](const TBreakPoint& bp)
-    //        {
-    //            jit->GetBlockCache()->InvalidateICache(bp.iAddress, 4);
-    //        }
-    //    );
-    //}
-
-    m_BreakPoints.clear();
-}

src/common/break_points.h

@@ -1,49 +0,0 @@
-// Copyright 2013 Dolphin Emulator Project / 2014 Citra Emulator Project
-// Licensed under GPLv2 or any later version
-// Refer to the license.txt file included.
-
-#pragma once
-
-#include <string>
-#include <vector>
-#include "common/common_types.h"
-
-class DebugInterface;
-
-struct TBreakPoint {
-    u32 iAddress;
-    bool bOn;
-    bool bTemporary;
-};
-
-// Code breakpoints.
-class BreakPoints {
-public:
-    typedef std::vector<TBreakPoint> TBreakPoints;
-    typedef std::vector<std::string> TBreakPointsStr;
-
-    const TBreakPoints& GetBreakPoints() {
-        return m_BreakPoints;
-    }
-
-    TBreakPointsStr GetStrings() const;
-    void AddFromStrings(const TBreakPointsStr& bps);
-
-    // is address breakpoint
-    bool IsAddressBreakPoint(u32 iAddress) const;
-    bool IsTempBreakPoint(u32 iAddress) const;
-
-    // Add BreakPoint
-    void Add(u32 em_address, bool temp = false);
-    void Add(const TBreakPoint& bp);
-
-    // Remove Breakpoint
-    void Remove(u32 iAddress);
-    void Clear();
-
-    void DeleteByAddress(u32 Address);
-
-private:
-    TBreakPoints m_BreakPoints;
-    u32 m_iBreakOnCount;
-};

src/common/color.h

@@ -4,6 +4,8 @@
 
 #pragma once
 
+#include <cstring>
+
 #include "common/common_types.h"
 #include "common/swap.h"
 #include "common/vector_math.h"
@@ -55,7 +57,7 @@ constexpr u8 Convert8To6(u8 value) {
  * @param bytes Pointer to encoded source color
  * @return Result color decoded as Math::Vec4<u8>
  */
-inline const Math::Vec4<u8> DecodeRGBA8(const u8* bytes) {
+inline Math::Vec4<u8> DecodeRGBA8(const u8* bytes) {
     return {bytes[3], bytes[2], bytes[1], bytes[0]};
 }
 
@@ -64,7 +66,7 @@ inline const Math::Vec4<u8> DecodeRGBA8(const u8* bytes) {
  * @param bytes Pointer to encoded source color
  * @return Result color decoded as Math::Vec4<u8>
  */
-inline const Math::Vec4<u8> DecodeRGB8(const u8* bytes) {
+inline Math::Vec4<u8> DecodeRGB8(const u8* bytes) {
     return {bytes[2], bytes[1], bytes[0], 255};
 }
 
@@ -73,7 +75,7 @@ inline const Math::Vec4<u8> DecodeRGB8(const u8* bytes) {
  * @param bytes Pointer to encoded source color
  * @return Result color decoded as Math::Vec4<u8>
  */
-inline const Math::Vec4<u8> DecodeRG8(const u8* bytes) {
+inline Math::Vec4<u8> DecodeRG8(const u8* bytes) {
     return {bytes[1], bytes[0], 0, 255};
 }
 
@@ -82,8 +84,9 @@ inline const Math::Vec4<u8> DecodeRG8(const u8* bytes) {
  * @param bytes Pointer to encoded source color
  * @return Result color decoded as Math::Vec4<u8>
  */
-inline const Math::Vec4<u8> DecodeRGB565(const u8* bytes) {
-    const u16_le pixel = *reinterpret_cast<const u16_le*>(bytes);
+inline Math::Vec4<u8> DecodeRGB565(const u8* bytes) {
+    u16_le pixel;
+    std::memcpy(&pixel, bytes, sizeof(pixel));
     return {Convert5To8((pixel >> 11) & 0x1F), Convert6To8((pixel >> 5) & 0x3F),
             Convert5To8(pixel & 0x1F), 255};
 }
@@ -93,8 +96,9 @@ inline const Math::Vec4<u8> DecodeRGB565(const u8* bytes) {
  * @param bytes Pointer to encoded source color
  * @return Result color decoded as Math::Vec4<u8>
  */
-inline const Math::Vec4<u8> DecodeRGB5A1(const u8* bytes) {
-    const u16_le pixel = *reinterpret_cast<const u16_le*>(bytes);
+inline Math::Vec4<u8> DecodeRGB5A1(const u8* bytes) {
+    u16_le pixel;
+    std::memcpy(&pixel, bytes, sizeof(pixel));
     return {Convert5To8((pixel >> 11) & 0x1F), Convert5To8((pixel >> 6) & 0x1F),
             Convert5To8((pixel >> 1) & 0x1F), Convert1To8(pixel & 0x1)};
 }
@@ -104,8 +108,9 @@ inline const Math::Vec4<u8> DecodeRGB5A1(const u8* bytes) {
  * @param bytes Pointer to encoded source color
  * @return Result color decoded as Math::Vec4<u8>
  */
-inline const Math::Vec4<u8> DecodeRGBA4(const u8* bytes) {
-    const u16_le pixel = *reinterpret_cast<const u16_le*>(bytes);
+inline Math::Vec4<u8> DecodeRGBA4(const u8* bytes) {
+    u16_le pixel;
+    std::memcpy(&pixel, bytes, sizeof(pixel));
     return {Convert4To8((pixel >> 12) & 0xF), Convert4To8((pixel >> 8) & 0xF),
             Convert4To8((pixel >> 4) & 0xF), Convert4To8(pixel & 0xF)};
 }
@@ -116,7 +121,9 @@ inline const Math::Vec4<u8> DecodeRGBA4(const u8* bytes) {
  * @return Depth value as an u32
  */
 inline u32 DecodeD16(const u8* bytes) {
-    return *reinterpret_cast<const u16_le*>(bytes);
+    u16_le data;
+    std::memcpy(&data, bytes, sizeof(data));
+    return data;
 }
 
 /**
@@ -133,7 +140,7 @@ inline u32 DecodeD24(const u8* bytes) {
  * @param bytes Pointer to encoded source values
  * @return Resulting values stored as a Math::Vec2
  */
-inline const Math::Vec2<u32> DecodeD24S8(const u8* bytes) {
+inline Math::Vec2<u32> DecodeD24S8(const u8* bytes) {
     return {static_cast<u32>((bytes[2] << 16) | (bytes[1] << 8) | bytes[0]), bytes[3]};
 }
 
@@ -175,8 +182,10 @@ inline void EncodeRG8(const Math::Vec4<u8>& color, u8* bytes) {
  * @param bytes Destination pointer to store encoded color
  */
 inline void EncodeRGB565(const Math::Vec4<u8>& color, u8* bytes) {
-    *reinterpret_cast<u16_le*>(bytes) =
+    const u16_le data =
         (Convert8To5(color.r()) << 11) | (Convert8To6(color.g()) << 5) | Convert8To5(color.b());
+
+    std::memcpy(bytes, &data, sizeof(data));
 }
 
 /**
@@ -185,9 +194,10 @@ inline void EncodeRGB565(const Math::Vec4<u8>& color, u8* bytes) {
  * @param bytes Destination pointer to store encoded color
  */
 inline void EncodeRGB5A1(const Math::Vec4<u8>& color, u8* bytes) {
-    *reinterpret_cast<u16_le*>(bytes) = (Convert8To5(color.r()) << 11) |
-                                        (Convert8To5(color.g()) << 6) |
-                                        (Convert8To5(color.b()) << 1) | Convert8To1(color.a());
+    const u16_le data = (Convert8To5(color.r()) << 11) | (Convert8To5(color.g()) << 6) |
+                        (Convert8To5(color.b()) << 1) | Convert8To1(color.a());
+
+    std::memcpy(bytes, &data, sizeof(data));
 }
 
 /**
@@ -196,9 +206,10 @@ inline void EncodeRGB5A1(const Math::Vec4<u8>& color, u8* bytes) {
  * @param bytes Destination pointer to store encoded color
  */
 inline void EncodeRGBA4(const Math::Vec4<u8>& color, u8* bytes) {
-    *reinterpret_cast<u16_le*>(bytes) = (Convert8To4(color.r()) << 12) |
-                                        (Convert8To4(color.g()) << 8) |
-                                        (Convert8To4(color.b()) << 4) | Convert8To4(color.a());
+    const u16 data = (Convert8To4(color.r()) << 12) | (Convert8To4(color.g()) << 8) |
+                     (Convert8To4(color.b()) << 4) | Convert8To4(color.a());
+
+    std::memcpy(bytes, &data, sizeof(data));
 }
 
 /**
@@ -207,7 +218,8 @@ inline void EncodeRGBA4(const Math::Vec4<u8>& color, u8* bytes) {
  * @param bytes Pointer where to store the encoded value
  */
 inline void EncodeD16(u32 value, u8* bytes) {
-    *reinterpret_cast<u16_le*>(bytes) = value & 0xFFFF;
+    const u16_le data = static_cast<u16>(value);
+    std::memcpy(bytes, &data, sizeof(data));
 }
 
 /**

src/common/common_funcs.h

@@ -6,7 +6,7 @@
 
 #include <string>
 
-#if !defined(ARCHITECTURE_x86_64) && !defined(ARCHITECTURE_ARM)
+#if !defined(ARCHITECTURE_x86_64)
 #include <cstdlib> // for exit
 #endif
 #include "common/common_types.h"
@@ -32,8 +32,6 @@
 
 #ifdef ARCHITECTURE_x86_64
 #define Crash() __asm__ __volatile__("int $3")
-#elif defined(ARCHITECTURE_ARM)
-#define Crash() __asm__ __volatile__("trap")
 #else
 #define Crash() exit(1)
 #endif

src/common/common_paths.h

@@ -26,20 +26,21 @@
 #define USA_DIR "USA"
 #define JAP_DIR "JAP"
 
-// Subdirs in the User dir returned by GetUserPath(D_USER_IDX)
+// Subdirs in the User dir returned by GetUserPath(UserPath::UserDir)
 #define CONFIG_DIR "config"
 #define CACHE_DIR "cache"
 #define SDMC_DIR "sdmc"
 #define NAND_DIR "nand"
 #define SYSDATA_DIR "sysdata"
+#define KEYS_DIR "keys"
 #define LOG_DIR "log"
 
 // Filenames
-// Files in the directory returned by GetUserPath(D_CONFIG_IDX)
+// Files in the directory returned by GetUserPath(UserPath::ConfigDir)
 #define EMU_CONFIG "emu.ini"
 #define DEBUGGER_CONFIG "debugger.ini"
 #define LOGGER_CONFIG "logger.ini"
-// Files in the directory returned by GetUserPath(D_LOGS_IDX)
+// Files in the directory returned by GetUserPath(UserPath::LogDir)
 #define LOG_FILE "yuzu_log.txt"
 
 // Sys files

src/common/file_util.cpp

@@ -2,7 +2,10 @@
 // Licensed under GPLv2 or any later version
 // Refer to the license.txt file included.
 
+#include <array>
+#include <memory>
 #include <sstream>
+#include <unordered_map>
 #include "common/assert.h"
 #include "common/common_funcs.h"
 #include "common/common_paths.h"
@@ -274,14 +277,10 @@ bool Copy(const std::string& srcFilename, const std::string& destFilename) {
               GetLastErrorMsg());
     return false;
 #else
-
-// buffer size
-#define BSIZE 1024
-
-    char buffer[BSIZE];
+    using CFilePointer = std::unique_ptr<FILE, decltype(&std::fclose)>;
 
     // Open input file
-    FILE* input = fopen(srcFilename.c_str(), "rb");
+    CFilePointer input{fopen(srcFilename.c_str(), "rb"), std::fclose};
     if (!input) {
         LOG_ERROR(Common_Filesystem, "opening input failed {} --> {}: {}", srcFilename,
                   destFilename, GetLastErrorMsg());
@@ -289,44 +288,36 @@ bool Copy(const std::string& srcFilename, const std::string& destFilename) {
     }
 
     // open output file
-    FILE* output = fopen(destFilename.c_str(), "wb");
+    CFilePointer output{fopen(destFilename.c_str(), "wb"), std::fclose};
     if (!output) {
-        fclose(input);
         LOG_ERROR(Common_Filesystem, "opening output failed {} --> {}: {}", srcFilename,
                   destFilename, GetLastErrorMsg());
         return false;
     }
 
     // copy loop
-    while (!feof(input)) {
+    std::array<char, 1024> buffer;
+    while (!feof(input.get())) {
         // read input
-        size_t rnum = fread(buffer, sizeof(char), BSIZE, input);
-        if (rnum != BSIZE) {
-            if (ferror(input) != 0) {
+        size_t rnum = fread(buffer.data(), sizeof(char), buffer.size(), input.get());
+        if (rnum != buffer.size()) {
+            if (ferror(input.get()) != 0) {
                 LOG_ERROR(Common_Filesystem, "failed reading from source, {} --> {}: {}",
                           srcFilename, destFilename, GetLastErrorMsg());
-                goto bail;
+                return false;
             }
         }
 
         // write output
-        size_t wnum = fwrite(buffer, sizeof(char), rnum, output);
+        size_t wnum = fwrite(buffer.data(), sizeof(char), rnum, output.get());
         if (wnum != rnum) {
             LOG_ERROR(Common_Filesystem, "failed writing to output, {} --> {}: {}", srcFilename,
                       destFilename, GetLastErrorMsg());
-            goto bail;
+            return false;
         }
     }
-    // close files
-    fclose(input);
-    fclose(output);
+
     return true;
-bail:
-    if (input)
-        fclose(input);
-    if (output)
-        fclose(output);
-    return false;
 #endif
 }
 
@@ -395,12 +386,12 @@ bool CreateEmptyFile(const std::string& filename) {
     return true;
 }
 
-bool ForeachDirectoryEntry(unsigned* num_entries_out, const std::string& directory,
+bool ForeachDirectoryEntry(u64* num_entries_out, const std::string& directory,
                            DirectoryEntryCallable callback) {
     LOG_TRACE(Common_Filesystem, "directory {}", directory);
 
     // How many files + directories we found
-    unsigned found_entries = 0;
+    u64 found_entries = 0;
 
     // Save the status of callback function
     bool callback_error = false;
@@ -430,7 +421,7 @@ bool ForeachDirectoryEntry(unsigned* num_entries_out, const std::string& directo
         if (virtual_name == "." || virtual_name == "..")
             continue;
 
-        unsigned ret_entries = 0;
+        u64 ret_entries = 0;
         if (!callback(&ret_entries, directory, virtual_name)) {
             callback_error = true;
             break;
@@ -454,9 +445,9 @@ bool ForeachDirectoryEntry(unsigned* num_entries_out, const std::string& directo
     return true;
 }
 
-unsigned ScanDirectoryTree(const std::string& directory, FSTEntry& parent_entry,
-                           unsigned int recursion) {
-    const auto callback = [recursion, &parent_entry](unsigned* num_entries_out,
+u64 ScanDirectoryTree(const std::string& directory, FSTEntry& parent_entry,
+                      unsigned int recursion) {
+    const auto callback = [recursion, &parent_entry](u64* num_entries_out,
                                                      const std::string& directory,
                                                      const std::string& virtual_name) -> bool {
         FSTEntry entry;
@@ -468,7 +459,7 @@ unsigned ScanDirectoryTree(const std::string& directory, FSTEntry& parent_entry,
             // is a directory, lets go inside if we didn't recurse to often
             if (recursion > 0) {
                 entry.size = ScanDirectoryTree(entry.physicalName, entry, recursion - 1);
-                *num_entries_out += (int)entry.size;
+                *num_entries_out += entry.size;
             } else {
                 entry.size = 0;
             }
@@ -479,16 +470,16 @@ unsigned ScanDirectoryTree(const std::string& directory, FSTEntry& parent_entry,
         (*num_entries_out)++;
 
         // Push into the tree
-        parent_entry.children.push_back(entry);
+        parent_entry.children.push_back(std::move(entry));
         return true;
     };
 
-    unsigned num_entries;
+    u64 num_entries;
     return ForeachDirectoryEntry(&num_entries, directory, callback) ? num_entries : 0;
 }
 
 bool DeleteDirRecursively(const std::string& directory, unsigned int recursion) {
-    const auto callback = [recursion](unsigned* num_entries_out, const std::string& directory,
+    const auto callback = [recursion](u64* num_entries_out, const std::string& directory,
                                       const std::string& virtual_name) -> bool {
         std::string new_path = directory + DIR_SEP_CHR + virtual_name;
 
@@ -681,67 +672,88 @@ std::string GetSysDirectory() {
 
 // Returns a string with a yuzu data dir or file in the user's home
 // directory. To be used in "multi-user" mode (that is, installed).
-const std::string& GetUserPath(const unsigned int DirIDX, const std::string& newPath) {
-    static std::string paths[NUM_PATH_INDICES];
+const std::string& GetUserPath(UserPath path, const std::string& new_path) {
+    static std::unordered_map<UserPath, std::string> paths;
+    auto& user_path = paths[UserPath::UserDir];
 
     // Set up all paths and files on the first run
-    if (paths[D_USER_IDX].empty()) {
+    if (user_path.empty()) {
 #ifdef _WIN32
-        paths[D_USER_IDX] = GetExeDirectory() + DIR_SEP USERDATA_DIR DIR_SEP;
-        if (!FileUtil::IsDirectory(paths[D_USER_IDX])) {
-            paths[D_USER_IDX] = AppDataRoamingDirectory() + DIR_SEP EMU_DATA_DIR DIR_SEP;
+        user_path = GetExeDirectory() + DIR_SEP USERDATA_DIR DIR_SEP;
+        if (!FileUtil::IsDirectory(user_path)) {
+            user_path = AppDataRoamingDirectory() + DIR_SEP EMU_DATA_DIR DIR_SEP;
         } else {
             LOG_INFO(Common_Filesystem, "Using the local user directory");
         }
 
-        paths[D_CONFIG_IDX] = paths[D_USER_IDX] + CONFIG_DIR DIR_SEP;
-        paths[D_CACHE_IDX] = paths[D_USER_IDX] + CACHE_DIR DIR_SEP;
+        paths.emplace(UserPath::ConfigDir, user_path + CONFIG_DIR DIR_SEP);
+        paths.emplace(UserPath::CacheDir, user_path + CACHE_DIR DIR_SEP);
 #else
         if (FileUtil::Exists(ROOT_DIR DIR_SEP USERDATA_DIR)) {
-            paths[D_USER_IDX] = ROOT_DIR DIR_SEP USERDATA_DIR DIR_SEP;
-            paths[D_CONFIG_IDX] = paths[D_USER_IDX] + CONFIG_DIR DIR_SEP;
-            paths[D_CACHE_IDX] = paths[D_USER_IDX] + CACHE_DIR DIR_SEP;
+            user_path = ROOT_DIR DIR_SEP USERDATA_DIR DIR_SEP;
+            paths.emplace(UserPath::ConfigDir, user_path + CONFIG_DIR DIR_SEP);
+            paths.emplace(UserPath::CacheDir, user_path + CACHE_DIR DIR_SEP);
         } else {
             std::string data_dir = GetUserDirectory("XDG_DATA_HOME");
             std::string config_dir = GetUserDirectory("XDG_CONFIG_HOME");
             std::string cache_dir = GetUserDirectory("XDG_CACHE_HOME");
 
-            paths[D_USER_IDX] = data_dir + DIR_SEP EMU_DATA_DIR DIR_SEP;
-            paths[D_CONFIG_IDX] = config_dir + DIR_SEP EMU_DATA_DIR DIR_SEP;
-            paths[D_CACHE_IDX] = cache_dir + DIR_SEP EMU_DATA_DIR DIR_SEP;
+            user_path = data_dir + DIR_SEP EMU_DATA_DIR DIR_SEP;
+            paths.emplace(UserPath::ConfigDir, config_dir + DIR_SEP EMU_DATA_DIR DIR_SEP);
+            paths.emplace(UserPath::CacheDir, cache_dir + DIR_SEP EMU_DATA_DIR DIR_SEP);
         }
 #endif
-        paths[D_SDMC_IDX] = paths[D_USER_IDX] + SDMC_DIR DIR_SEP;
-        paths[D_NAND_IDX] = paths[D_USER_IDX] + NAND_DIR DIR_SEP;
-        paths[D_SYSDATA_IDX] = paths[D_USER_IDX] + SYSDATA_DIR DIR_SEP;
+        paths.emplace(UserPath::SDMCDir, user_path + SDMC_DIR DIR_SEP);
+        paths.emplace(UserPath::NANDDir, user_path + NAND_DIR DIR_SEP);
+        paths.emplace(UserPath::SysDataDir, user_path + SYSDATA_DIR DIR_SEP);
+        paths.emplace(UserPath::KeysDir, user_path + KEYS_DIR DIR_SEP);
         // TODO: Put the logs in a better location for each OS
-        paths[D_LOGS_IDX] = paths[D_USER_IDX] + LOG_DIR DIR_SEP;
+        paths.emplace(UserPath::LogDir, user_path + LOG_DIR DIR_SEP);
     }
 
-    if (!newPath.empty()) {
-        if (!FileUtil::IsDirectory(newPath)) {
-            LOG_ERROR(Common_Filesystem, "Invalid path specified {}", newPath);
-            return paths[DirIDX];
+    if (!new_path.empty()) {
+        if (!FileUtil::IsDirectory(new_path)) {
+            LOG_ERROR(Common_Filesystem, "Invalid path specified {}", new_path);
+            return paths[path];
         } else {
-            paths[DirIDX] = newPath;
+            paths[path] = new_path;
         }
 
-        switch (DirIDX) {
-        case D_ROOT_IDX:
-            paths[D_USER_IDX] = paths[D_ROOT_IDX] + DIR_SEP;
+        switch (path) {
+        case UserPath::RootDir:
+            user_path = paths[UserPath::RootDir] + DIR_SEP;
             break;
 
-        case D_USER_IDX:
-            paths[D_USER_IDX] = paths[D_ROOT_IDX] + DIR_SEP;
-            paths[D_CONFIG_IDX] = paths[D_USER_IDX] + CONFIG_DIR DIR_SEP;
-            paths[D_CACHE_IDX] = paths[D_USER_IDX] + CACHE_DIR DIR_SEP;
-            paths[D_SDMC_IDX] = paths[D_USER_IDX] + SDMC_DIR DIR_SEP;
-            paths[D_NAND_IDX] = paths[D_USER_IDX] + NAND_DIR DIR_SEP;
+        case UserPath::UserDir:
+            user_path = paths[UserPath::RootDir] + DIR_SEP;
+            paths[UserPath::ConfigDir] = user_path + CONFIG_DIR DIR_SEP;
+            paths[UserPath::CacheDir] = user_path + CACHE_DIR DIR_SEP;
+            paths[UserPath::SDMCDir] = user_path + SDMC_DIR DIR_SEP;
+            paths[UserPath::NANDDir] = user_path + NAND_DIR DIR_SEP;
             break;
         }
     }
 
-    return paths[DirIDX];
+    return paths[path];
+}
+
+std::string GetHactoolConfigurationPath() {
+#ifdef _WIN32
+    PWSTR pw_local_path = nullptr;
+    if (SHGetKnownFolderPath(FOLDERID_Profile, 0, nullptr, &pw_local_path) != S_OK)
+        return "";
+    std::string local_path = Common::UTF16ToUTF8(pw_local_path);
+    CoTaskMemFree(pw_local_path);
+    return local_path + "\\.switch";
+#else
+    return GetHomeDirectory() + "/.switch";
+#endif
+}
+
+std::string GetNANDRegistrationDir(bool system) {
+    if (system)
+        return GetUserPath(UserPath::NANDDir) + "system/Contents/registered/";
+    return GetUserPath(UserPath::NANDDir) + "user/Contents/registered/";
 }
 
 size_t WriteStringToFile(bool text_file, const std::string& str, const char* filename) {
@@ -752,7 +764,7 @@ size_t ReadFileToString(bool text_file, const char* filename, std::string& str)
     IOFile file(filename, text_file ? "r" : "rb");
 
     if (!file.IsOpen())
-        return false;
+        return 0;
 
     str.resize(static_cast<u32>(file.GetSize()));
     return file.ReadArray(&str[0], str.size());
@@ -800,71 +812,103 @@ void SplitFilename83(const std::string& filename, std::array<char, 9>& short_nam
     }
 }
 
-std::vector<std::string> SplitPathComponents(const std::string& filename) {
-    auto copy(filename);
+std::vector<std::string> SplitPathComponents(std::string_view filename) {
+    std::string copy(filename);
     std::replace(copy.begin(), copy.end(), '\\', '/');
     std::vector<std::string> out;
 
-    std::stringstream stream(filename);
+    std::stringstream stream(copy);
     std::string item;
-    while (std::getline(stream, item, '/'))
+    while (std::getline(stream, item, '/')) {
         out.push_back(std::move(item));
+    }
 
     return out;
 }
 
-std::string GetParentPath(const std::string& path) {
-    auto out = path;
-    const auto name_bck_index = out.find_last_of('\\');
-    const auto name_fwd_index = out.find_last_of('/');
+std::string_view GetParentPath(std::string_view path) {
+    const auto name_bck_index = path.rfind('\\');
+    const auto name_fwd_index = path.rfind('/');
     size_t name_index;
-    if (name_bck_index == std::string::npos || name_fwd_index == std::string::npos)
-        name_index = std::min<size_t>(name_bck_index, name_fwd_index);
-    else
-        name_index = std::max<size_t>(name_bck_index, name_fwd_index);
 
-    return out.erase(name_index);
-}
-
-std::string GetPathWithoutTop(std::string path) {
-    if (path.empty())
-        return "";
-    while (path[0] == '\\' || path[0] == '/') {
-        path = path.substr(1);
-        if (path.empty())
-            return "";
+    if (name_bck_index == std::string_view::npos || name_fwd_index == std::string_view::npos) {
+        name_index = std::min(name_bck_index, name_fwd_index);
+    } else {
+        name_index = std::max(name_bck_index, name_fwd_index);
     }
-    const auto name_bck_index = path.find_first_of('\\');
-    const auto name_fwd_index = path.find_first_of('/');
-    return path.substr(std::min<size_t>(name_bck_index, name_fwd_index) + 1);
-    return path.substr(std::min<size_t>(name_bck_index, name_fwd_index) + 1);
+
+    return path.substr(0, name_index);
 }
 
-std::string GetFilename(std::string path) {
-    std::replace(path.begin(), path.end(), '\\', '/');
-    auto name_index = path.find_last_of('/');
-    if (name_index == std::string::npos)
-        return "";
+std::string_view GetPathWithoutTop(std::string_view path) {
+    if (path.empty()) {
+        return path;
+    }
+
+    while (path[0] == '\\' || path[0] == '/') {
+        path.remove_prefix(1);
+        if (path.empty()) {
+            return path;
+        }
+    }
+
+    const auto name_bck_index = path.find('\\');
+    const auto name_fwd_index = path.find('/');
+    return path.substr(std::min(name_bck_index, name_fwd_index) + 1);
+}
+
+std::string_view GetFilename(std::string_view path) {
+    const auto name_index = path.find_last_of("\\/");
+
+    if (name_index == std::string_view::npos) {
+        return {};
+    }
+
     return path.substr(name_index + 1);
 }
 
-std::string GetExtensionFromFilename(const std::string& name) {
-    size_t index = name.find_last_of('.');
-    if (index == std::string::npos)
-        return "";
+std::string_view GetExtensionFromFilename(std::string_view name) {
+    const size_t index = name.rfind('.');
+
+    if (index == std::string_view::npos) {
+        return {};
+    }
 
     return name.substr(index + 1);
 }
 
-std::string RemoveTrailingSlash(const std::string& path) {
-    if (path.empty())
+std::string_view RemoveTrailingSlash(std::string_view path) {
+    if (path.empty()) {
         return path;
-    if (path.back() == '\\' || path.back() == '/')
-        return path.substr(0, path.size() - 1);
+    }
+
+    if (path.back() == '\\' || path.back() == '/') {
+        path.remove_suffix(1);
+        return path;
+    }
 
     return path;
 }
 
+std::string SanitizePath(std::string_view path_, DirectorySeparator directory_separator) {
+    std::string path(path_);
+    char type1 = directory_separator == DirectorySeparator::BackwardSlash ? '/' : '\\';
+    char type2 = directory_separator == DirectorySeparator::BackwardSlash ? '\\' : '/';
+
+    if (directory_separator == DirectorySeparator::PlatformDefault) {
+#ifdef _WIN32
+        type1 = '/';
+        type2 = '\\';
+#endif
+    }
+
+    std::replace(path.begin(), path.end(), type1, type2);
+    path.erase(std::unique(path.begin(), path.end(),
+                           [type2](char c1, char c2) { return c1 == type2 && c2 == type2; }),
+               path.end());
+    return std::string(RemoveTrailingSlash(path));
+}
+
 IOFile::IOFile() {}
 
 IOFile::IOFile(const std::string& filename, const char openmode[], int flags) {

src/common/file_util.h

@@ -8,7 +8,9 @@
 #include <cstdio>
 #include <fstream>
 #include <functional>
+#include <limits>
 #include <string>
+#include <string_view>
 #include <type_traits>
 #include <vector>
 #include "common/common_types.h"
@@ -16,21 +18,21 @@
 #include "common/string_util.h"
 #endif
 
-// User directory indices for GetUserPath
-enum {
-    D_USER_IDX,
-    D_ROOT_IDX,
-    D_CONFIG_IDX,
-    D_CACHE_IDX,
-    D_SDMC_IDX,
-    D_NAND_IDX,
-    D_SYSDATA_IDX,
-    D_LOGS_IDX,
-    NUM_PATH_INDICES
-};
-
 namespace FileUtil {
 
+// User paths for GetUserPath
+enum class UserPath {
+    CacheDir,
+    ConfigDir,
+    KeysDir,
+    LogDir,
+    NANDDir,
+    RootDir,
+    SDMCDir,
+    SysDataDir,
+    UserDir,
+};
+
 // FileSystem tree node/
 struct FSTEntry {
     bool isDirectory;
@@ -85,7 +87,7 @@ bool CreateEmptyFile(const std::string& filename);
  * @return whether handling the entry succeeded
  */
 using DirectoryEntryCallable = std::function<bool(
-    unsigned* num_entries_out, const std::string& directory, const std::string& virtual_name)>;
+    u64* num_entries_out, const std::string& directory, const std::string& virtual_name)>;
 
 /**
  * Scans a directory, calling the callback for each file/directory contained within.
@@ -96,7 +98,7 @@ using DirectoryEntryCallable = std::function<bool(
  * @param callback The callback which will be called for each entry
  * @return whether scanning the directory succeeded
  */
-bool ForeachDirectoryEntry(unsigned* num_entries_out, const std::string& directory,
+bool ForeachDirectoryEntry(u64* num_entries_out, const std::string& directory,
                            DirectoryEntryCallable callback);
 
 /**
@@ -106,8 +108,8 @@ bool ForeachDirectoryEntry(unsigned* num_entries_out, const std::string& directo
  * @param recursion Number of children directories to read before giving up.
  * @return the total number of files/directories found
  */
-unsigned ScanDirectoryTree(const std::string& directory, FSTEntry& parent_entry,
-                           unsigned int recursion = 0);
+u64 ScanDirectoryTree(const std::string& directory, FSTEntry& parent_entry,
+                      unsigned int recursion = 0);
 
 // deletes the given directory and anything under it. Returns true on success.
 bool DeleteDirRecursively(const std::string& directory, unsigned int recursion = 256);
@@ -123,7 +125,11 @@ bool SetCurrentDir(const std::string& directory);
 
 // Returns a pointer to a string with a yuzu data dir in the user's home
 // directory. To be used in "multi-user" mode (that is, installed).
-const std::string& GetUserPath(const unsigned int DirIDX, const std::string& newPath = "");
+const std::string& GetUserPath(UserPath path, const std::string& new_path = "");
+
+std::string GetHactoolConfigurationPath();
+
+std::string GetNANDRegistrationDir(bool system = false);
 
 // Returns the path to where the sys file are
 std::string GetSysDirectory();
@@ -152,22 +158,22 @@ void SplitFilename83(const std::string& filename, std::array<char, 9>& short_nam
 
 // Splits the path on '/' or '\' and put the components into a vector
 // i.e. "C:\Users\Yuzu\Documents\save.bin" becomes {"C:", "Users", "Yuzu", "Documents", "save.bin" }
-std::vector<std::string> SplitPathComponents(const std::string& filename);
+std::vector<std::string> SplitPathComponents(std::string_view filename);
 
 // Gets all of the text up to the last '/' or '\' in the path.
-std::string GetParentPath(const std::string& path);
+std::string_view GetParentPath(std::string_view path);
 
 // Gets all of the text after the first '/' or '\' in the path.
-std::string GetPathWithoutTop(std::string path);
+std::string_view GetPathWithoutTop(std::string_view path);
 
 // Gets the filename of the path
-std::string GetFilename(std::string path);
+std::string_view GetFilename(std::string_view path);
 
 // Gets the extension of the filename
-std::string GetExtensionFromFilename(const std::string& name);
+std::string_view GetExtensionFromFilename(std::string_view name);
 
 // Removes the final '/' or '\' if one exists
-std::string RemoveTrailingSlash(const std::string& path);
+std::string_view RemoveTrailingSlash(std::string_view path);
 
 // Creates a new vector containing indices [first, last) from the original.
 template <typename T>
@@ -178,6 +184,13 @@ std::vector<T> SliceVector(const std::vector<T>& vector, size_t first, size_t la
     return std::vector<T>(vector.begin() + first, vector.begin() + first + last);
 }
 
+enum class DirectorySeparator { ForwardSlash, BackwardSlash, PlatformDefault };
+
+// Removes trailing slash, makes all '\\' into '/', and removes duplicate '/'. Makes '/' into '\\'
+// depending if directory_separator is BackwardSlash or PlatformDefault and running on windows
+std::string SanitizePath(std::string_view path,
+                         DirectorySeparator directory_separator = DirectorySeparator::ForwardSlash);
+
 // simple wrapper for cstdlib file functions to
 // hopefully will make error checking easier
 // and make forgetting an fclose() harder
@@ -201,39 +214,42 @@ public:
 
     template <typename T>
     size_t ReadArray(T* data, size_t length) const {
-        static_assert(std::is_trivially_copyable<T>(),
+        static_assert(std::is_trivially_copyable_v<T>,
                       "Given array does not consist of trivially copyable objects");
 
-        if (!IsOpen())
-            return -1;
+        if (!IsOpen()) {
+            return std::numeric_limits<size_t>::max();
+        }
 
         return std::fread(data, sizeof(T), length, m_file);
     }
 
     template <typename T>
     size_t WriteArray(const T* data, size_t length) {
-        static_assert(std::is_trivially_copyable<T>(),
+        static_assert(std::is_trivially_copyable_v<T>,
                       "Given array does not consist of trivially copyable objects");
-        if (!IsOpen())
-            return -1;
+        if (!IsOpen()) {
+            return std::numeric_limits<size_t>::max();
+        }
+
         return std::fwrite(data, sizeof(T), length, m_file);
     }
 
     template <typename T>
     size_t ReadBytes(T* data, size_t length) const {
-        static_assert(std::is_trivially_copyable<T>(), "T must be trivially copyable");
+        static_assert(std::is_trivially_copyable_v<T>, "T must be trivially copyable");
         return ReadArray(reinterpret_cast<char*>(data), length);
     }
 
     template <typename T>
     size_t WriteBytes(const T* data, size_t length) {
-        static_assert(std::is_trivially_copyable<T>(), "T must be trivially copyable");
+        static_assert(std::is_trivially_copyable_v<T>, "T must be trivially copyable");
         return WriteArray(reinterpret_cast<const char*>(data), length);
     }
 
     template <typename T>
     size_t WriteObject(const T& object) {
-        static_assert(!std::is_pointer<T>::value, "Given object is a pointer");
+        static_assert(!std::is_pointer_v<T>, "WriteObject arguments must not be a pointer");
         return WriteArray(&object, 1);
     }
 

src/common/hash.h

@@ -28,7 +28,7 @@ static inline u64 ComputeHash64(const void* data, size_t len) {
  */
 template <typename T>
 static inline u64 ComputeStructHash64(const T& data) {
-    static_assert(std::is_trivially_copyable<T>(),
+    static_assert(std::is_trivially_copyable_v<T>,
                   "Type passed to ComputeStructHash64 must be trivially copyable");
     return ComputeHash64(&data, sizeof(data));
 }
@@ -38,7 +38,7 @@ template <typename T>
 struct HashableStruct {
     // In addition to being trivially copyable, T must also have a trivial default constructor,
     // because any member initialization would be overridden by memset
-    static_assert(std::is_trivial<T>(), "Type passed to HashableStruct must be trivial");
+    static_assert(std::is_trivial_v<T>, "Type passed to HashableStruct must be trivial");
     /*
      * We use a union because "implicitly-defined copy/move constructor for a union X copies the
      * object representation of X." and "implicitly-defined copy assignment operator for a union X

src/common/hex_util.cpp

@@ -0,0 +1,43 @@
+// Copyright 2013 Dolphin Emulator Project / 2014 Citra Emulator Project
+// Licensed under GPLv2 or any later version
+// Refer to the license.txt file included.
+
+#include "common/hex_util.h"
+#include "common/logging/log.h"
+
+namespace Common {
+
+u8 ToHexNibble(char c1) {
+    if (c1 >= 65 && c1 <= 70)
+        return c1 - 55;
+    if (c1 >= 97 && c1 <= 102)
+        return c1 - 87;
+    if (c1 >= 48 && c1 <= 57)
+        return c1 - 48;
+    LOG_ERROR(Common, "Invalid hex digit: 0x{:02X}", c1);
+    return 0;
+}
+
+std::array<u8, 16> operator""_array16(const char* str, size_t len) {
+    if (len != 32) {
+        LOG_ERROR(Common,
+                  "Attempting to parse string to array that is not of correct size (expected=32, "
+                  "actual={}).",
+                  len);
+        return {};
+    }
+    return HexStringToArray<16>(str);
+}
+
+std::array<u8, 32> operator""_array32(const char* str, size_t len) {
+    if (len != 64) {
+        LOG_ERROR(Common,
+                  "Attempting to parse string to array that is not of correct size (expected=64, "
+                  "actual={}).",
+                  len);
+        return {};
+    }
+    return HexStringToArray<32>(str);
+}
+
+} // namespace Common

src/common/hex_util.h

@@ -0,0 +1,41 @@
+// Copyright 2013 Dolphin Emulator Project / 2014 Citra Emulator Project
+// Licensed under GPLv2 or any later version
+// Refer to the license.txt file included.
+
+#pragma once
+
+#include <array>
+#include <cstddef>
+#include <string>
+#include <fmt/format.h>
+#include "common/common_types.h"
+
+namespace Common {
+
+u8 ToHexNibble(char c1);
+
+template <size_t Size, bool le = false>
+std::array<u8, Size> HexStringToArray(std::string_view str) {
+    std::array<u8, Size> out{};
+    if constexpr (le) {
+        for (size_t i = 2 * Size - 2; i <= 2 * Size; i -= 2)
+            out[i / 2] = (ToHexNibble(str[i]) << 4) | ToHexNibble(str[i + 1]);
+    } else {
+        for (size_t i = 0; i < 2 * Size; i += 2)
+            out[i / 2] = (ToHexNibble(str[i]) << 4) | ToHexNibble(str[i + 1]);
+    }
+    return out;
+}
+
+template <size_t Size>
+std::string HexArrayToString(std::array<u8, Size> array, bool upper = true) {
+    std::string out;
+    for (u8 c : array)
+        out += fmt::format(upper ? "{:02X}" : "{:02x}", c);
+    return out;
+}
+
+std::array<u8, 0x10> operator"" _array16(const char* str, size_t len);
+std::array<u8, 0x20> operator"" _array32(const char* str, size_t len);
+
+} // namespace Common

src/common/logging/backend.cpp

@@ -3,19 +3,20 @@
 // Refer to the license.txt file included.
 
 #include <algorithm>
-#include <array>
+#include <atomic>
 #include <chrono>
 #include <climits>
 #include <condition_variable>
 #include <memory>
+#include <mutex>
 #include <thread>
+#include <vector>
 #ifdef _WIN32
 #include <share.h> // For _SH_DENYWR
 #else
 #define _SH_DENYWR 0
 #endif
 #include "common/assert.h"
-#include "common/common_funcs.h" // snprintf compatibility define
 #include "common/logging/backend.h"
 #include "common/logging/log.h"
 #include "common/logging/text_formatter.h"
@@ -48,11 +49,11 @@ public:
         backends.push_back(std::move(backend));
     }
 
-    void RemoveBackend(const std::string& backend_name) {
+    void RemoveBackend(std::string_view backend_name) {
         std::lock_guard<std::mutex> lock(writing_mutex);
-        auto it = std::remove_if(backends.begin(), backends.end(), [&backend_name](const auto& i) {
-            return !strcmp(i->GetName(), backend_name.c_str());
-        });
+        const auto it =
+            std::remove_if(backends.begin(), backends.end(),
+                           [&backend_name](const auto& i) { return backend_name == i->GetName(); });
         backends.erase(it, backends.end());
     }
 
@@ -64,10 +65,10 @@ public:
         filter = f;
     }
 
-    Backend* GetBackend(const std::string& backend_name) {
-        auto it = std::find_if(backends.begin(), backends.end(), [&backend_name](const auto& i) {
-            return !strcmp(i->GetName(), backend_name.c_str());
-        });
+    Backend* GetBackend(std::string_view backend_name) {
+        const auto it =
+            std::find_if(backends.begin(), backends.end(),
+                         [&backend_name](const auto& i) { return backend_name == i->GetName(); });
         if (it == backends.end())
             return nullptr;
         return it->get();
@@ -167,25 +168,49 @@ void FileBackend::Write(const Entry& entry) {
     SUB(Service, AM)                                                                               \
     SUB(Service, AOC)                                                                              \
     SUB(Service, APM)                                                                              \
+    SUB(Service, ARP)                                                                              \
     SUB(Service, BCAT)                                                                             \
+    SUB(Service, BPC)                                                                              \
+    SUB(Service, BTDRV)                                                                            \
+    SUB(Service, BTM)                                                                              \
+    SUB(Service, Capture)                                                                          \
+    SUB(Service, ERPT)                                                                             \
+    SUB(Service, ETicket)                                                                          \
+    SUB(Service, EUPLD)                                                                            \
     SUB(Service, Fatal)                                                                            \
+    SUB(Service, FGM)                                                                              \
     SUB(Service, Friend)                                                                           \
     SUB(Service, FS)                                                                               \
+    SUB(Service, GRC)                                                                              \
     SUB(Service, HID)                                                                              \
+    SUB(Service, LBL)                                                                              \
+    SUB(Service, LDN)                                                                              \
+    SUB(Service, LDR)                                                                              \
     SUB(Service, LM)                                                                               \
+    SUB(Service, Migration)                                                                        \
+    SUB(Service, Mii)                                                                              \
     SUB(Service, MM)                                                                               \
+    SUB(Service, NCM)                                                                              \
+    SUB(Service, NFC)                                                                              \
     SUB(Service, NFP)                                                                              \
     SUB(Service, NIFM)                                                                             \
+    SUB(Service, NIM)                                                                              \
     SUB(Service, NS)                                                                               \
     SUB(Service, NVDRV)                                                                            \
+    SUB(Service, PCIE)                                                                             \
     SUB(Service, PCTL)                                                                             \
+    SUB(Service, PCV)                                                                              \
+    SUB(Service, PM)                                                                               \
     SUB(Service, PREPO)                                                                            \
+    SUB(Service, PSC)                                                                              \
     SUB(Service, SET)                                                                              \
     SUB(Service, SM)                                                                               \
     SUB(Service, SPL)                                                                              \
     SUB(Service, SSL)                                                                              \
     SUB(Service, Time)                                                                             \
+    SUB(Service, USB)                                                                              \
     SUB(Service, VI)                                                                               \
+    SUB(Service, WLAN)                                                                             \
     CLS(HW)                                                                                        \
     SUB(HW, Memory)                                                                                \
     SUB(HW, LCD)                                                                                   \
@@ -202,6 +227,7 @@ void FileBackend::Write(const Entry& entry) {
     CLS(Input)                                                                                     \
     CLS(Network)                                                                                   \
     CLS(Loader)                                                                                    \
+    CLS(Crypto)                                                                                    \
     CLS(WebService)
 
 // GetClassName is a macro defined by Windows.h, grrr...
@@ -265,24 +291,25 @@ void AddBackend(std::unique_ptr<Backend> backend) {
     Impl::Instance().AddBackend(std::move(backend));
 }
 
-void RemoveBackend(const std::string& backend_name) {
+void RemoveBackend(std::string_view backend_name) {
     Impl::Instance().RemoveBackend(backend_name);
 }
 
-Backend* GetBackend(const std::string& backend_name) {
+Backend* GetBackend(std::string_view backend_name) {
     return Impl::Instance().GetBackend(backend_name);
 }
 
 void FmtLogMessageImpl(Class log_class, Level log_level, const char* filename,
                        unsigned int line_num, const char* function, const char* format,
                        const fmt::format_args& args) {
-    auto filter = Impl::Instance().GetGlobalFilter();
+    auto& instance = Impl::Instance();
+    const auto& filter = instance.GetGlobalFilter();
     if (!filter.CheckMessage(log_class, log_level))
         return;
 
     Entry entry =
         CreateEntry(log_class, log_level, filename, line_num, function, fmt::vformat(format, args));
 
-    Impl::Instance().PushEntry(std::move(entry));
+    instance.PushEntry(std::move(entry));
 }
 } // namespace Log

src/common/logging/backend.h

@@ -4,10 +4,9 @@
 #pragma once
 
 #include <chrono>
-#include <cstdarg>
 #include <memory>
 #include <string>
-#include <utility>
+#include <string_view>
 #include "common/file_util.h"
 #include "common/logging/filter.h"
 #include "common/logging/log.h"
@@ -106,9 +105,9 @@ private:
 
 void AddBackend(std::unique_ptr<Backend> backend);
 
-void RemoveBackend(const std::string& backend_name);
+void RemoveBackend(std::string_view backend_name);
 
-Backend* GetBackend(const std::string& backend_name);
+Backend* GetBackend(std::string_view backend_name);
 
 /**
  * Returns the name of the passed log class as a C-string. Subclasses are separated by periods

src/common/logging/filter.cpp

@@ -8,39 +8,9 @@
 #include "common/string_util.h"
 
 namespace Log {
-
-Filter::Filter(Level default_level) {
-    ResetAll(default_level);
-}
-
-void Filter::ResetAll(Level level) {
-    class_levels.fill(level);
-}
-
-void Filter::SetClassLevel(Class log_class, Level level) {
-    class_levels[static_cast<size_t>(log_class)] = level;
-}
-
-void Filter::ParseFilterString(const std::string& filter_str) {
-    auto clause_begin = filter_str.cbegin();
-    while (clause_begin != filter_str.cend()) {
-        auto clause_end = std::find(clause_begin, filter_str.cend(), ' ');
-
-        // If clause isn't empty
-        if (clause_end != clause_begin) {
-            ParseFilterRule(clause_begin, clause_end);
-        }
-
-        if (clause_end != filter_str.cend()) {
-            // Skip over the whitespace
-            ++clause_end;
-        }
-        clause_begin = clause_end;
-    }
-}
-
+namespace {
 template <typename It>
-static Level GetLevelByName(const It begin, const It end) {
+Level GetLevelByName(const It begin, const It end) {
     for (u8 i = 0; i < static_cast<u8>(Level::Count); ++i) {
         const char* level_name = GetLevelName(static_cast<Level>(i));
         if (Common::ComparePartialString(begin, end, level_name)) {
@@ -51,7 +21,7 @@ static Level GetLevelByName(const It begin, const It end) {
 }
 
 template <typename It>
-static Class GetClassByName(const It begin, const It end) {
+Class GetClassByName(const It begin, const It end) {
     for (ClassType i = 0; i < static_cast<ClassType>(Class::Count); ++i) {
         const char* level_name = GetLogClassName(static_cast<Class>(i));
         if (Common::ComparePartialString(begin, end, level_name)) {
@@ -61,8 +31,8 @@ static Class GetClassByName(const It begin, const It end) {
     return Class::Count;
 }
 
-bool Filter::ParseFilterRule(const std::string::const_iterator begin,
-                             const std::string::const_iterator end) {
+template <typename Iterator>
+bool ParseFilterRule(Filter& instance, Iterator begin, Iterator end) {
     auto level_separator = std::find(begin, end, ':');
     if (level_separator == end) {
         LOG_ERROR(Log, "Invalid log filter. Must specify a log level after `:`: {}",
@@ -77,7 +47,7 @@ bool Filter::ParseFilterRule(const std::string::const_iterator begin,
     }
 
     if (Common::ComparePartialString(begin, level_separator, "*")) {
-        ResetAll(level);
+        instance.ResetAll(level);
         return true;
     }
 
@@ -87,9 +57,40 @@ bool Filter::ParseFilterRule(const std::string::const_iterator begin,
         return false;
     }
 
-    SetClassLevel(log_class, level);
+    instance.SetClassLevel(log_class, level);
     return true;
 }
+} // Anonymous namespace
+
+Filter::Filter(Level default_level) {
+    ResetAll(default_level);
+}
+
+void Filter::ResetAll(Level level) {
+    class_levels.fill(level);
+}
+
+void Filter::SetClassLevel(Class log_class, Level level) {
+    class_levels[static_cast<size_t>(log_class)] = level;
+}
+
+void Filter::ParseFilterString(std::string_view filter_view) {
+    auto clause_begin = filter_view.cbegin();
+    while (clause_begin != filter_view.cend()) {
+        auto clause_end = std::find(clause_begin, filter_view.cend(), ' ');
+
+        // If clause isn't empty
+        if (clause_end != clause_begin) {
+            ParseFilterRule(*this, clause_begin, clause_end);
+        }
+
+        if (clause_end != filter_view.cend()) {
+            // Skip over the whitespace
+            ++clause_end;
+        }
+        clause_begin = clause_end;
+    }
+}
 
 bool Filter::CheckMessage(Class log_class, Level level) const {
     return static_cast<u8>(level) >= static_cast<u8>(class_levels[static_cast<size_t>(log_class)]);

src/common/logging/filter.h

@@ -6,7 +6,7 @@
 
 #include <array>
 #include <cstddef>
-#include <string>
+#include <string_view>
 #include "common/logging/log.h"
 
 namespace Log {
@@ -40,9 +40,7 @@ public:
      *  - `Service:Info` -- Sets the level of Service to Info.
      *  - `Service.FS:Trace` -- Sets the level of the Service.FS class to Trace.
      */
-    void ParseFilterString(const std::string& filter_str);
-    bool ParseFilterRule(const std::string::const_iterator start,
-                         const std::string::const_iterator end);
+    void ParseFilterString(std::string_view filter_view);
 
     /// Matches class/level combination against the filter, returning true if it passed.
     bool CheckMessage(Class log_class, Level level) const;

src/common/logging/log.h

@@ -54,26 +54,50 @@ enum class Class : ClassType {
     Service_AM,        ///< The AM (Applet manager) service
     Service_AOC,       ///< The AOC (AddOn Content) service
     Service_APM,       ///< The APM (Performance) service
+    Service_ARP,       ///< The ARP service
     Service_Audio,     ///< The Audio (Audio control) service
     Service_BCAT,      ///< The BCAT service
+    Service_BPC,       ///< The BPC service
+    Service_BTDRV,     ///< The Bluetooth driver service
+    Service_BTM,       ///< The BTM service
+    Service_Capture,   ///< The capture service
+    Service_ERPT,      ///< The error reporting service
+    Service_ETicket,   ///< The ETicket service
+    Service_EUPLD,     ///< The error upload service
     Service_Fatal,     ///< The Fatal service
+    Service_FGM,       ///< The FGM service
     Service_Friend,    ///< The friend service
     Service_FS,        ///< The FS (Filesystem) service
+    Service_GRC,       ///< The game recording service
     Service_HID,       ///< The HID (Human interface device) service
+    Service_LBL,       ///< The LBL (LCD backlight) service
+    Service_LDN,       ///< The LDN (Local domain network) service
+    Service_LDR,       ///< The loader service
     Service_LM,        ///< The LM (Logger) service
+    Service_Migration, ///< The migration service
+    Service_Mii,       ///< The Mii service
     Service_MM,        ///< The MM (Multimedia) service
+    Service_NCM,       ///< The NCM service
+    Service_NFC,       ///< The NFC (Near-field communication) service
     Service_NFP,       ///< The NFP service
     Service_NIFM,      ///< The NIFM (Network interface) service
+    Service_NIM,       ///< The NIM service
     Service_NS,        ///< The NS services
     Service_NVDRV,     ///< The NVDRV (Nvidia driver) service
+    Service_PCIE,      ///< The PCIe service
     Service_PCTL,      ///< The PCTL (Parental control) service
+    Service_PCV,       ///< The PCV service
+    Service_PM,        ///< The PM service
     Service_PREPO,     ///< The PREPO (Play report) service
+    Service_PSC,       ///< The PSC service
     Service_SET,       ///< The SET (Settings) service
     Service_SM,        ///< The SM (Service manager) service
     Service_SPL,       ///< The SPL service
     Service_SSL,       ///< The SSL service
     Service_Time,      ///< The time service
+    Service_USB,       ///< The USB (Universal Serial Bus) service
     Service_VI,        ///< The VI (Video interface) service
+    Service_WLAN,      ///< The WLAN (Wireless local area network) service
     HW,                ///< Low-level hardware emulation
     HW_Memory,         ///< Memory-map and address translation
     HW_LCD,            ///< LCD register emulation
@@ -88,6 +112,7 @@ enum class Class : ClassType {
     Audio_DSP,         ///< The HLE implementation of the DSP
     Audio_Sink,        ///< Emulator audio output backend
     Loader,            ///< ROM loader
+    Crypto,            ///< Cryptographic engine/functions
     Input,             ///< Input emulation
     Network,           ///< Network emulation
     WebService,        ///< Interface to yuzu Web Services

src/common/logging/text_formatter.cpp

@@ -42,7 +42,7 @@ void PrintColoredMessage(const Entry& entry) {
         return;
     }
 
-    CONSOLE_SCREEN_BUFFER_INFO original_info = {0};
+    CONSOLE_SCREEN_BUFFER_INFO original_info = {};
     GetConsoleScreenBufferInfo(console_handle, &original_info);
 
     WORD color = 0;

src/common/logging/text_formatter.h

@@ -5,6 +5,7 @@
 #pragma once
 
 #include <cstddef>
+#include <string>
 
 namespace Log {
 

src/common/math_util.h

@@ -19,12 +19,12 @@ inline bool IntervalsIntersect(unsigned start0, unsigned length0, unsigned start
 
 template <class T>
 struct Rectangle {
-    T left;
-    T top;
-    T right;
-    T bottom;
+    T left{};
+    T top{};
+    T right{};
+    T bottom{};
 
-    Rectangle() {}
+    Rectangle() = default;
 
     Rectangle(T left, T top, T right, T bottom)
         : left(left), top(top), right(right), bottom(bottom) {}

src/common/misc.cpp

@@ -4,7 +4,7 @@
 
 #include <cstddef>
 #ifdef _WIN32
-#include <Windows.h>
+#include <windows.h>
 #else
 #include <cerrno>
 #include <cstring>

src/common/string_util.cpp

@@ -2,12 +2,12 @@
 // Licensed under GPLv2 or any later version
 // Refer to the license.txt file included.
 
+#include <algorithm>
 #include <cctype>
 #include <cerrno>
 #include <cstdio>
 #include <cstdlib>
 #include <cstring>
-#include <boost/range/algorithm/transform.hpp>
 #include "common/common_paths.h"
 #include "common/logging/log.h"
 #include "common/string_util.h"
@@ -24,13 +24,15 @@ namespace Common {
 
 /// Make a string lowercase
 std::string ToLower(std::string str) {
-    boost::transform(str, str.begin(), ::tolower);
+    std::transform(str.begin(), str.end(), str.begin(),
+                   [](unsigned char c) { return std::tolower(c); });
     return str;
 }
 
 /// Make a string uppercase
 std::string ToUpper(std::string str) {
-    boost::transform(str, str.begin(), ::toupper);
+    std::transform(str.begin(), str.end(), str.begin(),
+                   [](unsigned char c) { return std::toupper(c); });
     return str;
 }
 
@@ -162,21 +164,21 @@ void SplitString(const std::string& str, const char delim, std::vector<std::stri
     std::istringstream iss(str);
     output.resize(1);
 
-    while (std::getline(iss, *output.rbegin(), delim))
-        output.push_back("");
+    while (std::getline(iss, *output.rbegin(), delim)) {
+        output.emplace_back();
+    }
 
     output.pop_back();
 }
 
-std::string TabsToSpaces(int tab_size, const std::string& in) {
-    const std::string spaces(tab_size, ' ');
-    std::string out(in);
-
+std::string TabsToSpaces(int tab_size, std::string in) {
     size_t i = 0;
-    while (out.npos != (i = out.find('\t')))
-        out.replace(i, 1, spaces);
 
-    return out;
+    while ((i = in.find('\t')) != std::string::npos) {
+        in.replace(i, 1, tab_size, ' ');
+    }
+
+    return in;
 }
 
 std::string ReplaceAll(std::string result, const std::string& src, const std::string& dest) {
@@ -220,31 +222,37 @@ std::u16string UTF8ToUTF16(const std::string& input) {
 }
 
 static std::wstring CPToUTF16(u32 code_page, const std::string& input) {
-    auto const size =
+    const auto size =
         MultiByteToWideChar(code_page, 0, input.data(), static_cast<int>(input.size()), nullptr, 0);
 
-    std::wstring output;
-    output.resize(size);
+    if (size == 0) {
+        return L"";
+    }
 
-    if (size == 0 ||
-        size != MultiByteToWideChar(code_page, 0, input.data(), static_cast<int>(input.size()),
-                                    &output[0], static_cast<int>(output.size())))
+    std::wstring output(size, L'\0');
+
+    if (size != MultiByteToWideChar(code_page, 0, input.data(), static_cast<int>(input.size()),
+                                    &output[0], static_cast<int>(output.size()))) {
         output.clear();
+    }
 
     return output;
 }
 
 std::string UTF16ToUTF8(const std::wstring& input) {
-    auto const size = WideCharToMultiByte(CP_UTF8, 0, input.data(), static_cast<int>(input.size()),
+    const auto size = WideCharToMultiByte(CP_UTF8, 0, input.data(), static_cast<int>(input.size()),
                                           nullptr, 0, nullptr, nullptr);
+    if (size == 0) {
+        return "";
+    }
 
-    std::string output;
-    output.resize(size);
+    std::string output(size, '\0');
 
-    if (size == 0 ||
-        size != WideCharToMultiByte(CP_UTF8, 0, input.data(), static_cast<int>(input.size()),
-                                    &output[0], static_cast<int>(output.size()), nullptr, nullptr))
+    if (size != WideCharToMultiByte(CP_UTF8, 0, input.data(), static_cast<int>(input.size()),
+                                    &output[0], static_cast<int>(output.size()), nullptr,
+                                    nullptr)) {
         output.clear();
+    }
 
     return output;
 }
@@ -265,8 +273,6 @@ std::string CP1252ToUTF8(const std::string& input) {
 
 template <typename T>
 static std::string CodeToUTF8(const char* fromcode, const std::basic_string<T>& input) {
-    std::string result;
-
     iconv_t const conv_desc = iconv_open("UTF-8", fromcode);
     if ((iconv_t)(-1) == conv_desc) {
         LOG_ERROR(Common, "Iconv initialization failure [{}]: {}", fromcode, strerror(errno));
@@ -278,8 +284,7 @@ static std::string CodeToUTF8(const char* fromcode, const std::basic_string<T>&
     // Multiply by 4, which is the max number of bytes to encode a codepoint
     const size_t out_buffer_size = 4 * in_bytes;
 
-    std::string out_buffer;
-    out_buffer.resize(out_buffer_size);
+    std::string out_buffer(out_buffer_size, '\0');
 
     auto src_buffer = &input[0];
     size_t src_bytes = in_bytes;
@@ -304,6 +309,7 @@ static std::string CodeToUTF8(const char* fromcode, const std::basic_string<T>&
         }
     }
 
+    std::string result;
     out_buffer.resize(out_buffer_size - dst_bytes);
     out_buffer.swap(result);
 
@@ -313,8 +319,6 @@ static std::string CodeToUTF8(const char* fromcode, const std::basic_string<T>&
 }
 
 std::u16string UTF8ToUTF16(const std::string& input) {
-    std::u16string result;
-
     iconv_t const conv_desc = iconv_open("UTF-16LE", "UTF-8");
     if ((iconv_t)(-1) == conv_desc) {
         LOG_ERROR(Common, "Iconv initialization failure [UTF-8]: {}", strerror(errno));
@@ -326,8 +330,7 @@ std::u16string UTF8ToUTF16(const std::string& input) {
     // Multiply by 4, which is the max number of bytes to encode a codepoint
     const size_t out_buffer_size = 4 * sizeof(char16_t) * in_bytes;
 
-    std::u16string out_buffer;
-    out_buffer.resize(out_buffer_size);
+    std::u16string out_buffer(out_buffer_size, char16_t{});
 
     char* src_buffer = const_cast<char*>(&input[0]);
     size_t src_bytes = in_bytes;
@@ -352,6 +355,7 @@ std::u16string UTF8ToUTF16(const std::string& input) {
         }
     }
 
+    std::u16string result;
     out_buffer.resize(out_buffer_size - dst_bytes);
     out_buffer.swap(result);
 

src/common/string_util.h

@@ -57,7 +57,7 @@ static bool TryParse(const std::string& str, N* const output) {
         return false;
 }
 
-std::string TabsToSpaces(int tab_size, const std::string& in);
+std::string TabsToSpaces(int tab_size, std::string in);
 
 void SplitString(const std::string& str, char delim, std::vector<std::string>& output);
 

src/common/swap.h

@@ -69,7 +69,7 @@ inline u32 swap32(u32 _data) {
 inline u64 swap64(u64 _data) {
     return _byteswap_uint64(_data);
 }
-#elif ARCHITECTURE_ARM
+#elif defined(ARCHITECTURE_ARM) && (__ARM_ARCH >= 6)
 inline u16 swap16(u16 _data) {
     u32 data = _data;
     __asm__("rev16 %0, %1\n" : "=l"(data) : "l"(data));

src/common/synchronized_wrapper.h

@@ -1,85 +0,0 @@
-// Copyright 2015 Citra Emulator Project
-// Licensed under GPLv2 or any later version
-// Refer to the license.txt file included.
-
-#pragma once
-
-#include <algorithm>
-#include <mutex>
-
-namespace Common {
-
-template <typename T>
-class SynchronizedWrapper;
-
-/**
- * Synchronized reference, that keeps a SynchronizedWrapper's mutex locked during its lifetime. This
- * greatly reduces the chance that someone will access the wrapped resource without locking the
- * mutex.
- */
-template <typename T>
-class SynchronizedRef {
-public:
-    SynchronizedRef(SynchronizedWrapper<T>& wrapper) : wrapper(&wrapper) {
-        wrapper.mutex.lock();
-    }
-
-    SynchronizedRef(SynchronizedRef&) = delete;
-    SynchronizedRef(SynchronizedRef&& o) : wrapper(o.wrapper) {
-        o.wrapper = nullptr;
-    }
-
-    ~SynchronizedRef() {
-        if (wrapper)
-            wrapper->mutex.unlock();
-    }
-
-    SynchronizedRef& operator=(SynchronizedRef&) = delete;
-    SynchronizedRef& operator=(SynchronizedRef&& o) {
-        std::swap(wrapper, o.wrapper);
-        return *this;
-    }
-
-    T& operator*() {
-        return wrapper->data;
-    }
-    const T& operator*() const {
-        return wrapper->data;
-    }
-
-    T* operator->() {
-        return &wrapper->data;
-    }
-    const T* operator->() const {
-        return &wrapper->data;
-    }
-
-private:
-    SynchronizedWrapper<T>* wrapper;
-};
-
-/**
- * Wraps an object, only allowing access to it via a locking reference wrapper. Good to ensure no
- * one forgets to lock a mutex before acessing an object. To access the wrapped object construct a
- * SyncronizedRef on this wrapper. Inspired by Rust's Mutex type
- * (http://doc.rust-lang.org/std/sync/struct.Mutex.html).
- */
-template <typename T>
-class SynchronizedWrapper {
-public:
-    template <typename... Args>
-    SynchronizedWrapper(Args&&... args) : data(std::forward<Args>(args)...) {}
-
-    SynchronizedRef<T> Lock() {
-        return {*this};
-    }
-
-private:
-    template <typename U>
-    friend class SynchronizedRef;
-
-    std::mutex mutex;
-    T data;
-};
-
-} // namespace Common

src/common/telemetry.cpp

@@ -3,8 +3,15 @@
 // Refer to the license.txt file included.
 
 #include <algorithm>
+#include <cstring>
+#include "common/assert.h"
+#include "common/scm_rev.h"
 #include "common/telemetry.h"
 
+#ifdef ARCHITECTURE_x86_64
+#include "common/x64/cpu_detect.h"
+#endif
+
 namespace Telemetry {
 
 void FieldCollection::Accept(VisitorInterface& visitor) const {
@@ -37,4 +44,62 @@ template class Field<std::string>;
 template class Field<const char*>;
 template class Field<std::chrono::microseconds>;
 
+#ifdef ARCHITECTURE_x86_64
+static const char* CpuVendorToStr(Common::CPUVendor vendor) {
+    switch (vendor) {
+    case Common::CPUVendor::INTEL:
+        return "Intel";
+    case Common::CPUVendor::AMD:
+        return "Amd";
+    case Common::CPUVendor::OTHER:
+        return "Other";
+    }
+    UNREACHABLE();
+}
+#endif
+
+void AppendBuildInfo(FieldCollection& fc) {
+    const bool is_git_dirty{std::strstr(Common::g_scm_desc, "dirty") != nullptr};
+    fc.AddField(FieldType::App, "Git_IsDirty", is_git_dirty);
+    fc.AddField(FieldType::App, "Git_Branch", Common::g_scm_branch);
+    fc.AddField(FieldType::App, "Git_Revision", Common::g_scm_rev);
+    fc.AddField(FieldType::App, "BuildDate", Common::g_build_date);
+    fc.AddField(FieldType::App, "BuildName", Common::g_build_name);
+}
+
+void AppendCPUInfo(FieldCollection& fc) {
+#ifdef ARCHITECTURE_x86_64
+    fc.AddField(FieldType::UserSystem, "CPU_Model", Common::GetCPUCaps().cpu_string);
+    fc.AddField(FieldType::UserSystem, "CPU_BrandString", Common::GetCPUCaps().brand_string);
+    fc.AddField(FieldType::UserSystem, "CPU_Vendor", CpuVendorToStr(Common::GetCPUCaps().vendor));
+    fc.AddField(FieldType::UserSystem, "CPU_Extension_x64_AES", Common::GetCPUCaps().aes);
+    fc.AddField(FieldType::UserSystem, "CPU_Extension_x64_AVX", Common::GetCPUCaps().avx);
+    fc.AddField(FieldType::UserSystem, "CPU_Extension_x64_AVX2", Common::GetCPUCaps().avx2);
+    fc.AddField(FieldType::UserSystem, "CPU_Extension_x64_BMI1", Common::GetCPUCaps().bmi1);
+    fc.AddField(FieldType::UserSystem, "CPU_Extension_x64_BMI2", Common::GetCPUCaps().bmi2);
+    fc.AddField(FieldType::UserSystem, "CPU_Extension_x64_FMA", Common::GetCPUCaps().fma);
+    fc.AddField(FieldType::UserSystem, "CPU_Extension_x64_FMA4", Common::GetCPUCaps().fma4);
+    fc.AddField(FieldType::UserSystem, "CPU_Extension_x64_SSE", Common::GetCPUCaps().sse);
+    fc.AddField(FieldType::UserSystem, "CPU_Extension_x64_SSE2", Common::GetCPUCaps().sse2);
+    fc.AddField(FieldType::UserSystem, "CPU_Extension_x64_SSE3", Common::GetCPUCaps().sse3);
+    fc.AddField(FieldType::UserSystem, "CPU_Extension_x64_SSSE3", Common::GetCPUCaps().ssse3);
+    fc.AddField(FieldType::UserSystem, "CPU_Extension_x64_SSE41", Common::GetCPUCaps().sse4_1);
+    fc.AddField(FieldType::UserSystem, "CPU_Extension_x64_SSE42", Common::GetCPUCaps().sse4_2);
+#else
+    fc.AddField(FieldType::UserSystem, "CPU_Model", "Other");
+#endif
+}
+
+void AppendOSInfo(FieldCollection& fc) {
+#ifdef __APPLE__
+    fc.AddField(FieldType::UserSystem, "OsPlatform", "Apple");
+#elif defined(_WIN32)
+    fc.AddField(FieldType::UserSystem, "OsPlatform", "Windows");
+#elif defined(__linux__) || defined(linux) || defined(__linux)
+    fc.AddField(FieldType::UserSystem, "OsPlatform", "Linux");
+#else
+    fc.AddField(FieldType::UserSystem, "OsPlatform", "Unknown");
+#endif
+}
+
 } // namespace Telemetry

src/common/telemetry.h

@@ -180,4 +180,16 @@ struct NullVisitor : public VisitorInterface {
     void Complete() override {}
 };
 
+/// Appends build-specific information to the given FieldCollection,
+/// such as branch name, revision hash, etc.
+void AppendBuildInfo(FieldCollection& fc);
+
+/// Appends CPU-specific information to the given FieldCollection,
+/// such as instruction set extensions, etc.
+void AppendCPUInfo(FieldCollection& fc);
+
+/// Appends OS-specific information to the given FieldCollection,
+/// such as platform name, etc.
+void AppendOSInfo(FieldCollection& fc);
+
 } // namespace Telemetry

src/common/thread_queue_list.h

@@ -16,7 +16,7 @@ struct ThreadQueueList {
     //               (dynamically resizable) circular buffers to remove their overhead when
     //               inserting and popping.
 
-    typedef unsigned int Priority;
+    using Priority = unsigned int;
 
     // Number of priority levels. (Valid levels are [0..NUM_QUEUES).)
     static const Priority NUM_QUEUES = N;
@@ -26,9 +26,9 @@ struct ThreadQueueList {
     }
 
     // Only for debugging, returns priority level.
-    Priority contains(const T& uid) {
+    Priority contains(const T& uid) const {
         for (Priority i = 0; i < NUM_QUEUES; ++i) {
-            Queue& cur = queues[i];
+            const Queue& cur = queues[i];
             if (std::find(cur.data.cbegin(), cur.data.cend(), uid) != cur.data.cend()) {
                 return i;
             }
@@ -37,8 +37,8 @@ struct ThreadQueueList {
         return -1;
     }
 
-    T get_first() {
-        Queue* cur = first;
+    T get_first() const {
+        const Queue* cur = first;
         while (cur != nullptr) {
             if (!cur->data.empty()) {
                 return cur->data.front();

src/common/threadsafe_queue.h

@@ -33,9 +33,11 @@ public:
     bool Empty() const {
         return !read_ptr->next.load();
     }
+
     T& Front() const {
         return read_ptr->current;
     }
+
     template <typename Arg>
     void Push(Arg&& t) {
         // create the element, add it to the queue
@@ -108,15 +110,41 @@ private:
 // single reader, multiple writer queue
 
 template <typename T, bool NeedSize = true>
-class MPSCQueue : public SPSCQueue<T, NeedSize> {
+class MPSCQueue {
 public:
+    u32 Size() const {
+        return spsc_queue.Size();
+    }
+
+    bool Empty() const {
+        return spsc_queue.Empty();
+    }
+
+    T& Front() const {
+        return spsc_queue.Front();
+    }
+
     template <typename Arg>
     void Push(Arg&& t) {
         std::lock_guard<std::mutex> lock(write_lock);
-        SPSCQueue<T, NeedSize>::Push(t);
+        spsc_queue.Push(t);
+    }
+
+    void Pop() {
+        return spsc_queue.Pop();
+    }
+
+    bool Pop(T& t) {
+        return spsc_queue.Pop(t);
+    }
+
+    // not thread-safe
+    void Clear() {
+        spsc_queue.Clear();
     }
 
 private:
+    SPSCQueue<T, NeedSize> spsc_queue;
     std::mutex write_lock;
 };
 } // namespace Common

src/common/timer.cpp

@@ -3,31 +3,16 @@
 // Refer to the license.txt file included.
 
 #include <ctime>
-
 #include <fmt/format.h>
-
-#ifdef _WIN32
-#include <windows.h>
-// windows.h needs to be included before other windows headers
-#include <mmsystem.h>
-#include <sys/timeb.h>
-#else
-#include <sys/time.h>
-#endif
 #include "common/common_types.h"
 #include "common/string_util.h"
 #include "common/timer.h"
 
 namespace Common {
 
-u32 Timer::GetTimeMs() {
-#ifdef _WIN32
-    return timeGetTime();
-#else
-    struct timeval t;
-    (void)gettimeofday(&t, nullptr);
-    return ((u32)(t.tv_sec * 1000 + t.tv_usec / 1000));
-#endif
+std::chrono::milliseconds Timer::GetTimeMs() {
+    return std::chrono::duration_cast<std::chrono::milliseconds>(
+        std::chrono::system_clock::now().time_since_epoch());
 }
 
 // --------------------------------------------
@@ -63,7 +48,7 @@ void Timer::Update() {
 // -------------------------------------
 
 // Get the number of milliseconds since the last Update()
-u64 Timer::GetTimeDifference() {
+std::chrono::milliseconds Timer::GetTimeDifference() {
     return GetTimeMs() - m_LastTime;
 }
 
@@ -74,11 +59,11 @@ void Timer::AddTimeDifference() {
 }
 
 // Get the time elapsed since the Start()
-u64 Timer::GetTimeElapsed() {
+std::chrono::milliseconds Timer::GetTimeElapsed() {
     // If we have not started yet, return 1 (because then I don't
     // have to change the FPS calculation in CoreRerecording.cpp .
-    if (m_StartTime == 0)
-        return 1;
+    if (m_StartTime.count() == 0)
+        return std::chrono::milliseconds(1);
 
     // Return the final timer time if the timer is stopped
     if (!m_Running)
@@ -90,49 +75,34 @@ u64 Timer::GetTimeElapsed() {
 // Get the formatted time elapsed since the Start()
 std::string Timer::GetTimeElapsedFormatted() const {
     // If we have not started yet, return zero
-    if (m_StartTime == 0)
+    if (m_StartTime.count() == 0)
         return "00:00:00:000";
 
     // The number of milliseconds since the start.
     // Use a different value if the timer is stopped.
-    u64 Milliseconds;
+    std::chrono::milliseconds Milliseconds;
     if (m_Running)
         Milliseconds = GetTimeMs() - m_StartTime;
     else
         Milliseconds = m_LastTime - m_StartTime;
     // Seconds
-    u32 Seconds = (u32)(Milliseconds / 1000);
+    std::chrono::seconds Seconds = std::chrono::duration_cast<std::chrono::seconds>(Milliseconds);
     // Minutes
-    u32 Minutes = Seconds / 60;
+    std::chrono::minutes Minutes = std::chrono::duration_cast<std::chrono::minutes>(Milliseconds);
     // Hours
-    u32 Hours = Minutes / 60;
+    std::chrono::hours Hours = std::chrono::duration_cast<std::chrono::hours>(Milliseconds);
 
-    std::string TmpStr = fmt::format("{:02}:{:02}:{:02}:{:03}", Hours, Minutes % 60, Seconds % 60,
-                                     Milliseconds % 1000);
+    std::string TmpStr = fmt::format("{:02}:{:02}:{:02}:{:03}", Hours.count(), Minutes.count() % 60,
+                                     Seconds.count() % 60, Milliseconds.count() % 1000);
     return TmpStr;
 }
 
-// Get current time
-void Timer::IncreaseResolution() {
-#ifdef _WIN32
-    timeBeginPeriod(1);
-#endif
-}
-
-void Timer::RestoreResolution() {
-#ifdef _WIN32
-    timeEndPeriod(1);
-#endif
-}
-
 // Get the number of seconds since January 1 1970
-u64 Timer::GetTimeSinceJan1970() {
-    time_t ltime;
-    time(&ltime);
-    return ((u64)ltime);
+std::chrono::seconds Timer::GetTimeSinceJan1970() {
+    return std::chrono::duration_cast<std::chrono::seconds>(GetTimeMs());
 }
 
-u64 Timer::GetLocalTimeSinceJan1970() {
+std::chrono::seconds Timer::GetLocalTimeSinceJan1970() {
     time_t sysTime, tzDiff, tzDST;
     struct tm* gmTime;
 
@@ -149,7 +119,7 @@ u64 Timer::GetLocalTimeSinceJan1970() {
     gmTime = gmtime(&sysTime);
     tzDiff = sysTime - mktime(gmTime);
 
-    return (u64)(sysTime + tzDiff + tzDST);
+    return std::chrono::seconds(sysTime + tzDiff + tzDST);
 }
 
 // Return the current time formatted as Minutes:Seconds:Milliseconds
@@ -164,30 +134,16 @@ std::string Timer::GetTimeFormatted() {
 
     strftime(tmp, 6, "%M:%S", gmTime);
 
-// Now tack on the milliseconds
-#ifdef _WIN32
-    struct timeb tp;
-    (void)::ftime(&tp);
-    return fmt::format("{}:{:03}", tmp, tp.millitm);
-#else
-    struct timeval t;
-    (void)gettimeofday(&t, nullptr);
-    return fmt::format("{}:{:03}", tmp, static_cast<int>(t.tv_usec / 1000));
-#endif
+    u64 milliseconds = static_cast<u64>(GetTimeMs().count()) % 1000;
+    return fmt::format("{}:{:03}", tmp, milliseconds);
 }
 
 // Returns a timestamp with decimals for precise time comparisons
 // ----------------
 double Timer::GetDoubleTime() {
-#ifdef _WIN32
-    struct timeb tp;
-    (void)::ftime(&tp);
-#else
-    struct timeval t;
-    (void)gettimeofday(&t, nullptr);
-#endif
     // Get continuous timestamp
-    u64 TmpSeconds = Common::Timer::GetTimeSinceJan1970();
+    u64 TmpSeconds = static_cast<u64>(Common::Timer::GetTimeSinceJan1970().count());
+    double ms = static_cast<u64>(GetTimeMs().count()) % 1000;
 
     // Remove a few years. We only really want enough seconds to make
     // sure that we are detecting actual actions, perhaps 60 seconds is
@@ -196,12 +152,7 @@ double Timer::GetDoubleTime() {
     TmpSeconds = TmpSeconds - (38 * 365 * 24 * 60 * 60);
 
     // Make a smaller integer that fits in the double
-    u32 Seconds = (u32)TmpSeconds;
-#ifdef _WIN32
-    double ms = tp.millitm / 1000.0 / 1000.0;
-#else
-    double ms = t.tv_usec / 1000000.0;
-#endif
+    u32 Seconds = static_cast<u32>(TmpSeconds);
     double TmpTime = Seconds + ms;
 
     return TmpTime;

src/common/timer.h

@@ -4,6 +4,7 @@
 
 #pragma once
 
+#include <chrono>
 #include <string>
 #include "common/common_types.h"
 
@@ -18,24 +19,22 @@ public:
 
     // The time difference is always returned in milliseconds, regardless of alternative internal
     // representation
-    u64 GetTimeDifference();
+    std::chrono::milliseconds GetTimeDifference();
     void AddTimeDifference();
 
-    static void IncreaseResolution();
-    static void RestoreResolution();
-    static u64 GetTimeSinceJan1970();
-    static u64 GetLocalTimeSinceJan1970();
+    static std::chrono::seconds GetTimeSinceJan1970();
+    static std::chrono::seconds GetLocalTimeSinceJan1970();
     static double GetDoubleTime();
 
     static std::string GetTimeFormatted();
     std::string GetTimeElapsedFormatted() const;
-    u64 GetTimeElapsed();
+    std::chrono::milliseconds GetTimeElapsed();
 
-    static u32 GetTimeMs();
+    static std::chrono::milliseconds GetTimeMs();
 
 private:
-    u64 m_LastTime;
-    u64 m_StartTime;
+    std::chrono::milliseconds m_LastTime;
+    std::chrono::milliseconds m_StartTime;
     bool m_Running;
 };
 

src/common/vector_math.h

@@ -42,140 +42,136 @@ class Vec3;
 template <typename T>
 class Vec4;
 
-template <typename T>
-static inline Vec2<T> MakeVec(const T& x, const T& y);
-template <typename T>
-static inline Vec3<T> MakeVec(const T& x, const T& y, const T& z);
-template <typename T>
-static inline Vec4<T> MakeVec(const T& x, const T& y, const T& z, const T& w);
-
 template <typename T>
 class Vec2 {
 public:
     T x{};
     T y{};
 
-    Vec2() = default;
-    Vec2(const T& _x, const T& _y) : x(_x), y(_y) {}
+    constexpr Vec2() = default;
+    constexpr Vec2(const T& x_, const T& y_) : x(x_), y(y_) {}
 
     template <typename T2>
-    Vec2<T2> Cast() const {
-        return Vec2<T2>((T2)x, (T2)y);
+    constexpr Vec2<T2> Cast() const {
+        return Vec2<T2>(static_cast<T2>(x), static_cast<T2>(y));
     }
 
-    static Vec2 AssignToAll(const T& f) {
-        return Vec2<T>(f, f);
+    static constexpr Vec2 AssignToAll(const T& f) {
+        return Vec2{f, f};
     }
 
-    Vec2<decltype(T{} + T{})> operator+(const Vec2& other) const {
-        return MakeVec(x + other.x, y + other.y);
+    constexpr Vec2<decltype(T{} + T{})> operator+(const Vec2& other) const {
+        return {x + other.x, y + other.y};
     }
-    void operator+=(const Vec2& other) {
+    constexpr Vec2& operator+=(const Vec2& other) {
         x += other.x;
         y += other.y;
+        return *this;
     }
-    Vec2<decltype(T{} - T{})> operator-(const Vec2& other) const {
-        return MakeVec(x - other.x, y - other.y);
+    constexpr Vec2<decltype(T{} - T{})> operator-(const Vec2& other) const {
+        return {x - other.x, y - other.y};
     }
-    void operator-=(const Vec2& other) {
+    constexpr Vec2& operator-=(const Vec2& other) {
         x -= other.x;
         y -= other.y;
+        return *this;
     }
 
     template <typename U = T>
-    Vec2<std::enable_if_t<std::is_signed<U>::value, U>> operator-() const {
-        return MakeVec(-x, -y);
+    constexpr Vec2<std::enable_if_t<std::is_signed_v<U>, U>> operator-() const {
+        return {-x, -y};
     }
-    Vec2<decltype(T{} * T{})> operator*(const Vec2& other) const {
-        return MakeVec(x * other.x, y * other.y);
-    }
-    template <typename V>
-    Vec2<decltype(T{} * V{})> operator*(const V& f) const {
-        return MakeVec(x * f, y * f);
-    }
-    template <typename V>
-    void operator*=(const V& f) {
-        *this = *this * f;
-    }
-    template <typename V>
-    Vec2<decltype(T{} / V{})> operator/(const V& f) const {
-        return MakeVec(x / f, y / f);
-    }
-    template <typename V>
-    void operator/=(const V& f) {
-        *this = *this / f;
+    constexpr Vec2<decltype(T{} * T{})> operator*(const Vec2& other) const {
+        return {x * other.x, y * other.y};
     }
 
-    T Length2() const {
+    template <typename V>
+    constexpr Vec2<decltype(T{} * V{})> operator*(const V& f) const {
+        return {x * f, y * f};
+    }
+
+    template <typename V>
+    constexpr Vec2& operator*=(const V& f) {
+        *this = *this * f;
+        return *this;
+    }
+
+    template <typename V>
+    constexpr Vec2<decltype(T{} / V{})> operator/(const V& f) const {
+        return {x / f, y / f};
+    }
+
+    template <typename V>
+    constexpr Vec2& operator/=(const V& f) {
+        *this = *this / f;
+        return *this;
+    }
+
+    constexpr T Length2() const {
         return x * x + y * y;
     }
 
     // Only implemented for T=float
     float Length() const;
-    void SetLength(const float l);
-    Vec2 WithLength(const float l) const;
-    float Distance2To(Vec2& other);
-    Vec2 Normalized() const;
     float Normalize(); // returns the previous length, which is often useful
 
-    T& operator[](int i) // allow vector[1] = 3   (vector.y=3)
-    {
+    constexpr T& operator[](std::size_t i) {
         return *((&x) + i);
     }
-    T operator[](const int i) const {
+    constexpr const T& operator[](std::size_t i) const {
         return *((&x) + i);
     }
 
-    void SetZero() {
+    constexpr void SetZero() {
         x = 0;
         y = 0;
     }
 
     // Common aliases: UV (texel coordinates), ST (texture coordinates)
-    T& u() {
+    constexpr T& u() {
         return x;
     }
-    T& v() {
+    constexpr T& v() {
         return y;
     }
-    T& s() {
+    constexpr T& s() {
         return x;
     }
-    T& t() {
+    constexpr T& t() {
         return y;
     }
 
-    const T& u() const {
+    constexpr const T& u() const {
         return x;
     }
-    const T& v() const {
+    constexpr const T& v() const {
         return y;
     }
-    const T& s() const {
+    constexpr const T& s() const {
         return x;
     }
-    const T& t() const {
+    constexpr const T& t() const {
         return y;
     }
 
     // swizzlers - create a subvector of specific components
-    const Vec2 yx() const {
+    constexpr Vec2 yx() const {
         return Vec2(y, x);
     }
-    const Vec2 vu() const {
+    constexpr Vec2 vu() const {
         return Vec2(y, x);
     }
-    const Vec2 ts() const {
+    constexpr Vec2 ts() const {
         return Vec2(y, x);
     }
 };
 
 template <typename T, typename V>
-Vec2<T> operator*(const V& f, const Vec2<T>& vec) {
+constexpr Vec2<T> operator*(const V& f, const Vec2<T>& vec) {
     return Vec2<T>(f * vec.x, f * vec.y);
 }
 
-typedef Vec2<float> Vec2f;
+using Vec2f = Vec2<float>;
 
 template <>
 inline float Vec2<float>::Length() const {
@@ -196,147 +192,151 @@ public:
     T y{};
     T z{};
 
-    Vec3() = default;
-    Vec3(const T& _x, const T& _y, const T& _z) : x(_x), y(_y), z(_z) {}
+    constexpr Vec3() = default;
+    constexpr Vec3(const T& x_, const T& y_, const T& z_) : x(x_), y(y_), z(z_) {}
 
     template <typename T2>
-    Vec3<T2> Cast() const {
-        return MakeVec<T2>((T2)x, (T2)y, (T2)z);
+    constexpr Vec3<T2> Cast() const {
+        return Vec3<T2>(static_cast<T2>(x), static_cast<T2>(y), static_cast<T2>(z));
     }
 
-    // Only implemented for T=int and T=float
-    static Vec3 FromRGB(unsigned int rgb);
-    unsigned int ToRGB() const; // alpha bits set to zero
-
-    static Vec3 AssignToAll(const T& f) {
-        return MakeVec(f, f, f);
+    static constexpr Vec3 AssignToAll(const T& f) {
+        return Vec3(f, f, f);
     }
 
-    Vec3<decltype(T{} + T{})> operator+(const Vec3& other) const {
-        return MakeVec(x + other.x, y + other.y, z + other.z);
+    constexpr Vec3<decltype(T{} + T{})> operator+(const Vec3& other) const {
+        return {x + other.x, y + other.y, z + other.z};
     }
-    void operator+=(const Vec3& other) {
+
+    constexpr Vec3& operator+=(const Vec3& other) {
         x += other.x;
         y += other.y;
         z += other.z;
+        return *this;
     }
-    Vec3<decltype(T{} - T{})> operator-(const Vec3& other) const {
-        return MakeVec(x - other.x, y - other.y, z - other.z);
+
+    constexpr Vec3<decltype(T{} - T{})> operator-(const Vec3& other) const {
+        return {x - other.x, y - other.y, z - other.z};
     }
-    void operator-=(const Vec3& other) {
+
+    constexpr Vec3& operator-=(const Vec3& other) {
         x -= other.x;
         y -= other.y;
         z -= other.z;
+        return *this;
     }
 
     template <typename U = T>
-    Vec3<std::enable_if_t<std::is_signed<U>::value, U>> operator-() const {
-        return MakeVec(-x, -y, -z);
-    }
-    Vec3<decltype(T{} * T{})> operator*(const Vec3& other) const {
-        return MakeVec(x * other.x, y * other.y, z * other.z);
-    }
-    template <typename V>
-    Vec3<decltype(T{} * V{})> operator*(const V& f) const {
-        return MakeVec(x * f, y * f, z * f);
-    }
-    template <typename V>
-    void operator*=(const V& f) {
-        *this = *this * f;
-    }
-    template <typename V>
-    Vec3<decltype(T{} / V{})> operator/(const V& f) const {
-        return MakeVec(x / f, y / f, z / f);
-    }
-    template <typename V>
-    void operator/=(const V& f) {
-        *this = *this / f;
+    constexpr Vec3<std::enable_if_t<std::is_signed_v<U>, U>> operator-() const {
+        return {-x, -y, -z};
     }
 
-    T Length2() const {
+    constexpr Vec3<decltype(T{} * T{})> operator*(const Vec3& other) const {
+        return {x * other.x, y * other.y, z * other.z};
+    }
+
+    template <typename V>
+    constexpr Vec3<decltype(T{} * V{})> operator*(const V& f) const {
+        return {x * f, y * f, z * f};
+    }
+
+    template <typename V>
+    constexpr Vec3& operator*=(const V& f) {
+        *this = *this * f;
+        return *this;
+    }
+    template <typename V>
+    constexpr Vec3<decltype(T{} / V{})> operator/(const V& f) const {
+        return {x / f, y / f, z / f};
+    }
+
+    template <typename V>
+    constexpr Vec3& operator/=(const V& f) {
+        *this = *this / f;
+        return *this;
+    }
+
+    constexpr T Length2() const {
         return x * x + y * y + z * z;
     }
 
     // Only implemented for T=float
     float Length() const;
-    void SetLength(const float l);
-    Vec3 WithLength(const float l) const;
-    float Distance2To(Vec3& other);
     Vec3 Normalized() const;
     float Normalize(); // returns the previous length, which is often useful
 
-    T& operator[](int i) // allow vector[2] = 3   (vector.z=3)
-    {
-        return *((&x) + i);
-    }
-    T operator[](const int i) const {
+    constexpr T& operator[](std::size_t i) {
         return *((&x) + i);
     }
 
-    void SetZero() {
+    constexpr const T& operator[](std::size_t i) const {
+        return *((&x) + i);
+    }
+
+    constexpr void SetZero() {
         x = 0;
         y = 0;
         z = 0;
     }
 
     // Common aliases: UVW (texel coordinates), RGB (colors), STQ (texture coordinates)
-    T& u() {
+    constexpr T& u() {
         return x;
     }
-    T& v() {
+    constexpr T& v() {
         return y;
     }
-    T& w() {
+    constexpr T& w() {
         return z;
     }
 
-    T& r() {
+    constexpr T& r() {
         return x;
     }
-    T& g() {
+    constexpr T& g() {
         return y;
     }
-    T& b() {
+    constexpr T& b() {
         return z;
     }
 
-    T& s() {
+    constexpr T& s() {
         return x;
     }
-    T& t() {
+    constexpr T& t() {
         return y;
     }
-    T& q() {
+    constexpr T& q() {
         return z;
     }
 
-    const T& u() const {
+    constexpr const T& u() const {
         return x;
     }
-    const T& v() const {
+    constexpr const T& v() const {
         return y;
     }
-    const T& w() const {
+    constexpr const T& w() const {
         return z;
     }
 
-    const T& r() const {
+    constexpr const T& r() const {
         return x;
     }
-    const T& g() const {
+    constexpr const T& g() const {
         return y;
     }
-    const T& b() const {
+    constexpr const T& b() const {
         return z;
     }
 
-    const T& s() const {
+    constexpr const T& s() const {
         return x;
     }
-    const T& t() const {
+    constexpr const T& t() const {
         return y;
     }
-    const T& q() const {
+    constexpr const T& q() const {
         return z;
     }
 
@@ -345,7 +345,7 @@ public:
 // _DEFINE_SWIZZLER2 defines a single such function, DEFINE_SWIZZLER2 defines all of them for all
 // component names (x<->r) and permutations (xy<->yx)
 #define _DEFINE_SWIZZLER2(a, b, name)                                                              \
-    const Vec2<T> name() const {                                                                   \
+    constexpr Vec2<T> name() const {                                                               \
         return Vec2<T>(a, b);                                                                      \
     }
 #define DEFINE_SWIZZLER2(a, b, a2, b2, a3, b3, a4, b4)                                             \
@@ -366,7 +366,7 @@ public:
 };
 
 template <typename T, typename V>
-Vec3<T> operator*(const V& f, const Vec3<T>& vec) {
+constexpr Vec3<T> operator*(const V& f, const Vec3<T>& vec) {
     return Vec3<T>(f * vec.x, f * vec.y, f * vec.z);
 }
 
@@ -387,7 +387,7 @@ inline float Vec3<float>::Normalize() {
     return length;
 }
 
-typedef Vec3<float> Vec3f;
+using Vec3f = Vec3<float>;
 
 template <typename T>
 class Vec4 {
@@ -397,86 +397,88 @@ public:
     T z{};
     T w{};
 
-    Vec4() = default;
-    Vec4(const T& _x, const T& _y, const T& _z, const T& _w) : x(_x), y(_y), z(_z), w(_w) {}
+    constexpr Vec4() = default;
+    constexpr Vec4(const T& x_, const T& y_, const T& z_, const T& w_)
+        : x(x_), y(y_), z(z_), w(w_) {}
 
     template <typename T2>
-    Vec4<T2> Cast() const {
-        return Vec4<T2>((T2)x, (T2)y, (T2)z, (T2)w);
+    constexpr Vec4<T2> Cast() const {
+        return Vec4<T2>(static_cast<T2>(x), static_cast<T2>(y), static_cast<T2>(z),
+                        static_cast<T2>(w));
     }
 
-    // Only implemented for T=int and T=float
-    static Vec4 FromRGBA(unsigned int rgba);
-    unsigned int ToRGBA() const;
-
-    static Vec4 AssignToAll(const T& f) {
-        return Vec4<T>(f, f, f, f);
+    static constexpr Vec4 AssignToAll(const T& f) {
+        return Vec4(f, f, f, f);
     }
 
-    Vec4<decltype(T{} + T{})> operator+(const Vec4& other) const {
-        return MakeVec(x + other.x, y + other.y, z + other.z, w + other.w);
+    constexpr Vec4<decltype(T{} + T{})> operator+(const Vec4& other) const {
+        return {x + other.x, y + other.y, z + other.z, w + other.w};
     }
-    void operator+=(const Vec4& other) {
+
+    constexpr Vec4& operator+=(const Vec4& other) {
         x += other.x;
         y += other.y;
         z += other.z;
         w += other.w;
+        return *this;
     }
-    Vec4<decltype(T{} - T{})> operator-(const Vec4& other) const {
-        return MakeVec(x - other.x, y - other.y, z - other.z, w - other.w);
+
+    constexpr Vec4<decltype(T{} - T{})> operator-(const Vec4& other) const {
+        return {x - other.x, y - other.y, z - other.z, w - other.w};
     }
-    void operator-=(const Vec4& other) {
+
+    constexpr Vec4& operator-=(const Vec4& other) {
         x -= other.x;
         y -= other.y;
         z -= other.z;
         w -= other.w;
+        return *this;
     }
 
     template <typename U = T>
-    Vec4<std::enable_if_t<std::is_signed<U>::value, U>> operator-() const {
-        return MakeVec(-x, -y, -z, -w);
-    }
-    Vec4<decltype(T{} * T{})> operator*(const Vec4& other) const {
-        return MakeVec(x * other.x, y * other.y, z * other.z, w * other.w);
-    }
-    template <typename V>
-    Vec4<decltype(T{} * V{})> operator*(const V& f) const {
-        return MakeVec(x * f, y * f, z * f, w * f);
-    }
-    template <typename V>
-    void operator*=(const V& f) {
-        *this = *this * f;
-    }
-    template <typename V>
-    Vec4<decltype(T{} / V{})> operator/(const V& f) const {
-        return MakeVec(x / f, y / f, z / f, w / f);
-    }
-    template <typename V>
-    void operator/=(const V& f) {
-        *this = *this / f;
+    constexpr Vec4<std::enable_if_t<std::is_signed_v<U>, U>> operator-() const {
+        return {-x, -y, -z, -w};
     }
 
-    T Length2() const {
+    constexpr Vec4<decltype(T{} * T{})> operator*(const Vec4& other) const {
+        return {x * other.x, y * other.y, z * other.z, w * other.w};
+    }
+
+    template <typename V>
+    constexpr Vec4<decltype(T{} * V{})> operator*(const V& f) const {
+        return {x * f, y * f, z * f, w * f};
+    }
+
+    template <typename V>
+    constexpr Vec4& operator*=(const V& f) {
+        *this = *this * f;
+        return *this;
+    }
+
+    template <typename V>
+    constexpr Vec4<decltype(T{} / V{})> operator/(const V& f) const {
+        return {x / f, y / f, z / f, w / f};
+    }
+
+    template <typename V>
+    constexpr Vec4& operator/=(const V& f) {
+        *this = *this / f;
+        return *this;
+    }
+
+    constexpr T Length2() const {
         return x * x + y * y + z * z + w * w;
     }
 
-    // Only implemented for T=float
-    float Length() const;
-    void SetLength(const float l);
-    Vec4 WithLength(const float l) const;
-    float Distance2To(Vec4& other);
-    Vec4 Normalized() const;
-    float Normalize(); // returns the previous length, which is often useful
-
-    T& operator[](int i) // allow vector[2] = 3   (vector.z=3)
-    {
-        return *((&x) + i);
-    }
-    T operator[](const int i) const {
+    constexpr T& operator[](std::size_t i) {
         return *((&x) + i);
     }
 
-    void SetZero() {
+    constexpr const T& operator[](std::size_t i) const {
+        return *((&x) + i);
+    }
+
+    constexpr void SetZero() {
         x = 0;
         y = 0;
         z = 0;
@@ -484,29 +486,29 @@ public:
     }
 
     // Common alias: RGBA (colors)
-    T& r() {
+    constexpr T& r() {
         return x;
     }
-    T& g() {
+    constexpr T& g() {
         return y;
     }
-    T& b() {
+    constexpr T& b() {
         return z;
     }
-    T& a() {
+    constexpr T& a() {
         return w;
     }
 
-    const T& r() const {
+    constexpr const T& r() const {
         return x;
     }
-    const T& g() const {
+    constexpr const T& g() const {
         return y;
     }
-    const T& b() const {
+    constexpr const T& b() const {
         return z;
     }
-    const T& a() const {
+    constexpr const T& a() const {
         return w;
     }
 
@@ -518,7 +520,7 @@ public:
 // DEFINE_SWIZZLER2_COMP2 defines two component functions for all component names (x<->r) and
 // permutations (xy<->yx)
 #define _DEFINE_SWIZZLER2(a, b, name)                                                              \
-    const Vec2<T> name() const {                                                                   \
+    constexpr Vec2<T> name() const {                                                               \
         return Vec2<T>(a, b);                                                                      \
     }
 #define DEFINE_SWIZZLER2_COMP1(a, a2)                                                              \
@@ -545,7 +547,7 @@ public:
 #undef _DEFINE_SWIZZLER2
 
 #define _DEFINE_SWIZZLER3(a, b, c, name)                                                           \
-    const Vec3<T> name() const {                                                                   \
+    constexpr Vec3<T> name() const {                                                               \
         return Vec3<T>(a, b, c);                                                                   \
     }
 #define DEFINE_SWIZZLER3_COMP1(a, a2)                                                              \
@@ -579,51 +581,51 @@ public:
 };
 
 template <typename T, typename V>
-Vec4<decltype(V{} * T{})> operator*(const V& f, const Vec4<T>& vec) {
-    return MakeVec(f * vec.x, f * vec.y, f * vec.z, f * vec.w);
+constexpr Vec4<decltype(V{} * T{})> operator*(const V& f, const Vec4<T>& vec) {
+    return {f * vec.x, f * vec.y, f * vec.z, f * vec.w};
 }
 
-typedef Vec4<float> Vec4f;
+using Vec4f = Vec4<float>;
 
 template <typename T>
-static inline decltype(T{} * T{} + T{} * T{}) Dot(const Vec2<T>& a, const Vec2<T>& b) {
+constexpr decltype(T{} * T{} + T{} * T{}) Dot(const Vec2<T>& a, const Vec2<T>& b) {
     return a.x * b.x + a.y * b.y;
 }
 
 template <typename T>
-static inline decltype(T{} * T{} + T{} * T{}) Dot(const Vec3<T>& a, const Vec3<T>& b) {
+constexpr decltype(T{} * T{} + T{} * T{}) Dot(const Vec3<T>& a, const Vec3<T>& b) {
     return a.x * b.x + a.y * b.y + a.z * b.z;
 }
 
 template <typename T>
-static inline decltype(T{} * T{} + T{} * T{}) Dot(const Vec4<T>& a, const Vec4<T>& b) {
+constexpr decltype(T{} * T{} + T{} * T{}) Dot(const Vec4<T>& a, const Vec4<T>& b) {
     return a.x * b.x + a.y * b.y + a.z * b.z + a.w * b.w;
 }
 
 template <typename T>
-static inline Vec3<decltype(T{} * T{} - T{} * T{})> Cross(const Vec3<T>& a, const Vec3<T>& b) {
-    return MakeVec(a.y * b.z - a.z * b.y, a.z * b.x - a.x * b.z, a.x * b.y - a.y * b.x);
+constexpr Vec3<decltype(T{} * T{} - T{} * T{})> Cross(const Vec3<T>& a, const Vec3<T>& b) {
+    return {a.y * b.z - a.z * b.y, a.z * b.x - a.x * b.z, a.x * b.y - a.y * b.x};
 }
 
 // linear interpolation via float: 0.0=begin, 1.0=end
 template <typename X>
-static inline decltype(X{} * float{} + X{} * float{}) Lerp(const X& begin, const X& end,
-                                                           const float t) {
+constexpr decltype(X{} * float{} + X{} * float{}) Lerp(const X& begin, const X& end,
+                                                       const float t) {
     return begin * (1.f - t) + end * t;
 }
 
 // linear interpolation via int: 0=begin, base=end
 template <typename X, int base>
-static inline decltype((X{} * int{} + X{} * int{}) / base) LerpInt(const X& begin, const X& end,
-                                                                   const int t) {
+constexpr decltype((X{} * int{} + X{} * int{}) / base) LerpInt(const X& begin, const X& end,
+                                                               const int t) {
     return (begin * (base - t) + end * t) / base;
 }
 
 // bilinear interpolation. s is for interpolating x00-x01 and x10-x11, and t is for the second
 // interpolation.
 template <typename X>
-inline auto BilinearInterp(const X& x00, const X& x01, const X& x10, const X& x11, const float s,
-                           const float t) {
+constexpr auto BilinearInterp(const X& x00, const X& x01, const X& x10, const X& x11, const float s,
+                              const float t) {
     auto y0 = Lerp(x00, x01, s);
     auto y1 = Lerp(x10, x11, s);
     return Lerp(y0, y1, t);
@@ -631,42 +633,42 @@ inline auto BilinearInterp(const X& x00, const X& x01, const X& x10, const X& x1
 
 // Utility vector factories
 template <typename T>
-static inline Vec2<T> MakeVec(const T& x, const T& y) {
+constexpr Vec2<T> MakeVec(const T& x, const T& y) {
     return Vec2<T>{x, y};
 }
 
 template <typename T>
-static inline Vec3<T> MakeVec(const T& x, const T& y, const T& z) {
+constexpr Vec3<T> MakeVec(const T& x, const T& y, const T& z) {
     return Vec3<T>{x, y, z};
 }
 
 template <typename T>
-static inline Vec4<T> MakeVec(const T& x, const T& y, const Vec2<T>& zw) {
+constexpr Vec4<T> MakeVec(const T& x, const T& y, const Vec2<T>& zw) {
     return MakeVec(x, y, zw[0], zw[1]);
 }
 
 template <typename T>
-static inline Vec3<T> MakeVec(const Vec2<T>& xy, const T& z) {
+constexpr Vec3<T> MakeVec(const Vec2<T>& xy, const T& z) {
     return MakeVec(xy[0], xy[1], z);
 }
 
 template <typename T>
-static inline Vec3<T> MakeVec(const T& x, const Vec2<T>& yz) {
+constexpr Vec3<T> MakeVec(const T& x, const Vec2<T>& yz) {
     return MakeVec(x, yz[0], yz[1]);
 }
 
 template <typename T>
-static inline Vec4<T> MakeVec(const T& x, const T& y, const T& z, const T& w) {
+constexpr Vec4<T> MakeVec(const T& x, const T& y, const T& z, const T& w) {
     return Vec4<T>{x, y, z, w};
 }
 
 template <typename T>
-static inline Vec4<T> MakeVec(const Vec2<T>& xy, const T& z, const T& w) {
+constexpr Vec4<T> MakeVec(const Vec2<T>& xy, const T& z, const T& w) {
     return MakeVec(xy[0], xy[1], z, w);
 }
 
 template <typename T>
-static inline Vec4<T> MakeVec(const T& x, const Vec2<T>& yz, const T& w) {
+constexpr Vec4<T> MakeVec(const T& x, const Vec2<T>& yz, const T& w) {
     return MakeVec(x, yz[0], yz[1], w);
 }
 
@@ -674,17 +676,17 @@ static inline Vec4<T> MakeVec(const T& x, const Vec2<T>& yz, const T& w) {
 //       Even if someone wanted to use an odd object like Vec2<Vec2<T>>, the compiler would error
 //       out soon enough due to misuse of the returned structure.
 template <typename T>
-static inline Vec4<T> MakeVec(const Vec2<T>& xy, const Vec2<T>& zw) {
+constexpr Vec4<T> MakeVec(const Vec2<T>& xy, const Vec2<T>& zw) {
     return MakeVec(xy[0], xy[1], zw[0], zw[1]);
 }
 
 template <typename T>
-static inline Vec4<T> MakeVec(const Vec3<T>& xyz, const T& w) {
+constexpr Vec4<T> MakeVec(const Vec3<T>& xyz, const T& w) {
     return MakeVec(xyz[0], xyz[1], xyz[2], w);
 }
 
 template <typename T>
-static inline Vec4<T> MakeVec(const T& x, const Vec3<T>& yzw) {
+constexpr Vec4<T> MakeVec(const T& x, const Vec3<T>& yzw) {
     return MakeVec(x, yzw[0], yzw[1], yzw[2]);
 }
 

src/common/x64/xbyak_abi.h

@@ -9,10 +9,9 @@
 #include "common/assert.h"
 #include "common/bit_set.h"
 
-namespace Common {
-namespace X64 {
+namespace Common::X64 {
 
-int RegToIndex(const Xbyak::Reg& reg) {
+inline int RegToIndex(const Xbyak::Reg& reg) {
     using Kind = Xbyak::Reg::Kind;
     ASSERT_MSG((reg.getKind() & (Kind::REG | Kind::XMM)) != 0,
                "RegSet only support GPRs and XMM registers.");
@@ -152,8 +151,8 @@ constexpr size_t ABI_SHADOW_SPACE = 0;
 
 #endif
 
-void ABI_CalculateFrameSize(BitSet32 regs, size_t rsp_alignment, size_t needed_frame_size,
-                            s32* out_subtraction, s32* out_xmm_offset) {
+inline void ABI_CalculateFrameSize(BitSet32 regs, size_t rsp_alignment, size_t needed_frame_size,
+                                   s32* out_subtraction, s32* out_xmm_offset) {
     int count = (regs & ABI_ALL_GPRS).Count();
     rsp_alignment -= count * 8;
     size_t subtraction = 0;
@@ -174,8 +173,8 @@ void ABI_CalculateFrameSize(BitSet32 regs, size_t rsp_alignment, size_t needed_f
     *out_xmm_offset = (s32)(subtraction - xmm_base_subtraction);
 }
 
-size_t ABI_PushRegistersAndAdjustStack(Xbyak::CodeGenerator& code, BitSet32 regs,
-                                       size_t rsp_alignment, size_t needed_frame_size = 0) {
+inline size_t ABI_PushRegistersAndAdjustStack(Xbyak::CodeGenerator& code, BitSet32 regs,
+                                              size_t rsp_alignment, size_t needed_frame_size = 0) {
     s32 subtraction, xmm_offset;
     ABI_CalculateFrameSize(regs, rsp_alignment, needed_frame_size, &subtraction, &xmm_offset);
 
@@ -195,8 +194,8 @@ size_t ABI_PushRegistersAndAdjustStack(Xbyak::CodeGenerator& code, BitSet32 regs
     return ABI_SHADOW_SPACE;
 }
 
-void ABI_PopRegistersAndAdjustStack(Xbyak::CodeGenerator& code, BitSet32 regs, size_t rsp_alignment,
-                                    size_t needed_frame_size = 0) {
+inline void ABI_PopRegistersAndAdjustStack(Xbyak::CodeGenerator& code, BitSet32 regs,
+                                           size_t rsp_alignment, size_t needed_frame_size = 0) {
     s32 subtraction, xmm_offset;
     ABI_CalculateFrameSize(regs, rsp_alignment, needed_frame_size, &subtraction, &xmm_offset);
 
@@ -217,5 +216,4 @@ void ABI_PopRegistersAndAdjustStack(Xbyak::CodeGenerator& code, BitSet32 regs, s
     }
 }
 
-} // namespace X64
-} // namespace Common
+} // namespace Common::X64

src/common/x64/xbyak_util.h

@@ -8,8 +8,7 @@
 #include <xbyak.h>
 #include "common/x64/xbyak_abi.h"
 
-namespace Common {
-namespace X64 {
+namespace Common::X64 {
 
 // Constants for use with cmpps/cmpss
 enum {
@@ -34,7 +33,7 @@ inline bool IsWithin2G(const Xbyak::CodeGenerator& code, uintptr_t target) {
 
 template <typename T>
 inline void CallFarFunction(Xbyak::CodeGenerator& code, const T f) {
-    static_assert(std::is_pointer<T>(), "Argument must be a (function) pointer.");
+    static_assert(std::is_pointer_v<T>, "Argument must be a (function) pointer.");
     size_t addr = reinterpret_cast<size_t>(f);
     if (IsWithin2G(code, addr)) {
         code.call(f);
@@ -45,5 +44,4 @@ inline void CallFarFunction(Xbyak::CodeGenerator& code, const T f) {
     }
 }
 
-} // namespace X64
-} // namespace Common
+} // namespace Common::X64

src/core/CMakeLists.txt

@@ -1,5 +1,7 @@
 add_library(core STATIC
     arm/arm_interface.h
+    arm/exclusive_monitor.cpp
+    arm/exclusive_monitor.h
     arm/unicorn/arm_unicorn.cpp
     arm/unicorn/arm_unicorn.h
     core.cpp
@@ -8,15 +10,39 @@ add_library(core STATIC
     core_cpu.h
     core_timing.cpp
     core_timing.h
+    core_timing_util.cpp
+    core_timing_util.h
+    crypto/aes_util.cpp
+    crypto/aes_util.h
+    crypto/encryption_layer.cpp
+    crypto/encryption_layer.h
+    crypto/key_manager.cpp
+    crypto/key_manager.h
+    crypto/ctr_encryption_layer.cpp
+    crypto/ctr_encryption_layer.h
+    crypto/xts_encryption_layer.cpp
+    crypto/xts_encryption_layer.h
+    file_sys/bis_factory.cpp
+    file_sys/bis_factory.h
+    file_sys/card_image.cpp
+    file_sys/card_image.h
     file_sys/content_archive.cpp
     file_sys/content_archive.h
+    file_sys/control_metadata.cpp
+    file_sys/control_metadata.h
     file_sys/directory.h
     file_sys/errors.h
     file_sys/mode.h
+    file_sys/nca_metadata.cpp
+    file_sys/nca_metadata.h
     file_sys/partition_filesystem.cpp
     file_sys/partition_filesystem.h
     file_sys/program_metadata.cpp
     file_sys/program_metadata.h
+    file_sys/registered_cache.cpp
+    file_sys/registered_cache.h
+    file_sys/romfs.cpp
+    file_sys/romfs.h
     file_sys/romfs_factory.cpp
     file_sys/romfs_factory.h
     file_sys/savedata_factory.cpp
@@ -25,10 +51,16 @@ add_library(core STATIC
     file_sys/sdmc_factory.h
     file_sys/vfs.cpp
     file_sys/vfs.h
+    file_sys/vfs_concat.cpp
+    file_sys/vfs_concat.h
     file_sys/vfs_offset.cpp
     file_sys/vfs_offset.h
     file_sys/vfs_real.cpp
     file_sys/vfs_real.h
+    file_sys/vfs_vector.cpp
+    file_sys/vfs_vector.h
+    file_sys/xts_archive.cpp
+    file_sys/xts_archive.h
     frontend/emu_window.cpp
     frontend/emu_window.h
     frontend/framebuffer_layout.cpp
@@ -36,8 +68,6 @@ add_library(core STATIC
     frontend/input.h
     gdbstub/gdbstub.cpp
     gdbstub/gdbstub.h
-    hle/config_mem.cpp
-    hle/config_mem.h
     hle/ipc.h
     hle/ipc_helpers.h
     hle/kernel/address_arbiter.cpp
@@ -55,12 +85,10 @@ add_library(core STATIC
     hle/kernel/hle_ipc.h
     hle/kernel/kernel.cpp
     hle/kernel/kernel.h
-    hle/kernel/memory.cpp
-    hle/kernel/memory.h
     hle/kernel/mutex.cpp
     hle/kernel/mutex.h
-    hle/kernel/object_address_table.cpp
-    hle/kernel/object_address_table.h
+    hle/kernel/object.cpp
+    hle/kernel/object.h
     hle/kernel/process.cpp
     hle/kernel/process.h
     hle/kernel/resource_limit.cpp
@@ -88,8 +116,6 @@ add_library(core STATIC
     hle/lock.cpp
     hle/lock.h
     hle/result.h
-    hle/romfs.cpp
-    hle/romfs.h
     hle/service/acc/acc.cpp
     hle/service/acc/acc.h
     hle/service/acc/acc_aa.cpp
@@ -100,38 +126,72 @@ add_library(core STATIC
     hle/service/acc/acc_u0.h
     hle/service/acc/acc_u1.cpp
     hle/service/acc/acc_u1.h
+    hle/service/acc/profile_manager.cpp
+    hle/service/acc/profile_manager.h
     hle/service/am/am.cpp
     hle/service/am/am.h
     hle/service/am/applet_ae.cpp
     hle/service/am/applet_ae.h
     hle/service/am/applet_oe.cpp
     hle/service/am/applet_oe.h
+    hle/service/am/idle.cpp
+    hle/service/am/idle.h
+    hle/service/am/omm.cpp
+    hle/service/am/omm.h
+    hle/service/am/spsm.cpp
+    hle/service/am/spsm.h
     hle/service/aoc/aoc_u.cpp
     hle/service/aoc/aoc_u.h
     hle/service/apm/apm.cpp
     hle/service/apm/apm.h
     hle/service/apm/interface.cpp
     hle/service/apm/interface.h
-    hle/service/audio/audio.cpp
-    hle/service/audio/audio.h
+    hle/service/arp/arp.cpp
+    hle/service/arp/arp.h
+    hle/service/audio/audctl.cpp
+    hle/service/audio/audctl.h
+    hle/service/audio/auddbg.cpp
+    hle/service/audio/auddbg.h
+    hle/service/audio/audin_a.cpp
+    hle/service/audio/audin_a.h
     hle/service/audio/audin_u.cpp
     hle/service/audio/audin_u.h
+    hle/service/audio/audio.cpp
+    hle/service/audio/audio.h
+    hle/service/audio/audout_a.cpp
+    hle/service/audio/audout_a.h
     hle/service/audio/audout_u.cpp
     hle/service/audio/audout_u.h
+    hle/service/audio/audrec_a.cpp
+    hle/service/audio/audrec_a.h
     hle/service/audio/audrec_u.cpp
     hle/service/audio/audrec_u.h
-    hle/service/audio/audren_u.cpp
-    hle/service/audio/audren_u.h
+    hle/service/audio/audren_a.cpp
+    hle/service/audio/audren_a.h
     hle/service/audio/audren_u.cpp
     hle/service/audio/audren_u.h
     hle/service/audio/codecctl.cpp
     hle/service/audio/codecctl.h
     hle/service/audio/hwopus.cpp
     hle/service/audio/hwopus.h
-    hle/service/bcat/module.cpp
-    hle/service/bcat/module.h
     hle/service/bcat/bcat.cpp
     hle/service/bcat/bcat.h
+    hle/service/bcat/module.cpp
+    hle/service/bcat/module.h
+    hle/service/bpc/bpc.cpp
+    hle/service/bpc/bpc.h
+    hle/service/btdrv/btdrv.cpp
+    hle/service/btdrv/btdrv.h
+    hle/service/btm/btm.cpp
+    hle/service/btm/btm.h
+    hle/service/caps/caps.cpp
+    hle/service/caps/caps.h
+    hle/service/erpt/erpt.cpp
+    hle/service/erpt/erpt.h
+    hle/service/es/es.cpp
+    hle/service/es/es.h
+    hle/service/eupld/eupld.cpp
+    hle/service/eupld/eupld.h
     hle/service/fatal/fatal.cpp
     hle/service/fatal/fatal.h
     hle/service/fatal/fatal_p.cpp
@@ -140,32 +200,52 @@ add_library(core STATIC
     hle/service/fatal/fatal_u.h
     hle/service/filesystem/filesystem.cpp
     hle/service/filesystem/filesystem.h
+    hle/service/filesystem/fsp_ldr.cpp
+    hle/service/filesystem/fsp_ldr.h
+    hle/service/filesystem/fsp_pr.cpp
+    hle/service/filesystem/fsp_pr.h
     hle/service/filesystem/fsp_srv.cpp
     hle/service/filesystem/fsp_srv.h
+    hle/service/fgm/fgm.cpp
+    hle/service/fgm/fgm.h
     hle/service/friend/friend.cpp
     hle/service/friend/friend.h
-    hle/service/friend/friend_a.cpp
-    hle/service/friend/friend_a.h
-    hle/service/friend/friend_u.cpp
-    hle/service/friend/friend_u.h
+    hle/service/friend/interface.cpp
+    hle/service/friend/interface.h
+    hle/service/grc/grc.cpp
+    hle/service/grc/grc.h
     hle/service/hid/hid.cpp
     hle/service/hid/hid.h
+    hle/service/hid/irs.cpp
+    hle/service/hid/irs.h
+    hle/service/hid/xcd.cpp
+    hle/service/hid/xcd.h
+    hle/service/lbl/lbl.cpp
+    hle/service/lbl/lbl.h
+    hle/service/ldn/ldn.cpp
+    hle/service/ldn/ldn.h
+    hle/service/ldr/ldr.cpp
+    hle/service/ldr/ldr.h
     hle/service/lm/lm.cpp
     hle/service/lm/lm.h
+    hle/service/mig/mig.cpp
+    hle/service/mig/mig.h
+    hle/service/mii/mii.cpp
+    hle/service/mii/mii.h
     hle/service/mm/mm_u.cpp
     hle/service/mm/mm_u.h
-    hle/service/nifm/nifm.cpp
-    hle/service/nifm/nifm.h
-    hle/service/nifm/nifm_a.cpp
-    hle/service/nifm/nifm_a.h
-    hle/service/nifm/nifm_s.cpp
-    hle/service/nifm/nifm_s.h
-    hle/service/nifm/nifm_u.cpp
-    hle/service/nifm/nifm_u.h
+    hle/service/ncm/ncm.cpp
+    hle/service/ncm/ncm.h
+    hle/service/nfc/nfc.cpp
+    hle/service/nfc/nfc.h
     hle/service/nfp/nfp.cpp
     hle/service/nfp/nfp.h
     hle/service/nfp/nfp_user.cpp
     hle/service/nfp/nfp_user.h
+    hle/service/nifm/nifm.cpp
+    hle/service/nifm/nifm.h
+    hle/service/nim/nim.cpp
+    hle/service/nim/nim.h
     hle/service/ns/ns.cpp
     hle/service/ns/ns.h
     hle/service/ns/pl_u.cpp
@@ -183,6 +263,10 @@ add_library(core STATIC
     hle/service/nvdrv/devices/nvhost_gpu.h
     hle/service/nvdrv/devices/nvhost_nvdec.cpp
     hle/service/nvdrv/devices/nvhost_nvdec.h
+    hle/service/nvdrv/devices/nvhost_nvjpg.cpp
+    hle/service/nvdrv/devices/nvhost_nvjpg.h
+    hle/service/nvdrv/devices/nvhost_vic.cpp
+    hle/service/nvdrv/devices/nvhost_vic.h
     hle/service/nvdrv/devices/nvmap.cpp
     hle/service/nvdrv/devices/nvmap.h
     hle/service/nvdrv/interface.cpp
@@ -195,12 +279,20 @@ add_library(core STATIC
     hle/service/nvflinger/buffer_queue.h
     hle/service/nvflinger/nvflinger.cpp
     hle/service/nvflinger/nvflinger.h
+    hle/service/pcie/pcie.cpp
+    hle/service/pcie/pcie.h
     hle/service/pctl/module.cpp
     hle/service/pctl/module.h
     hle/service/pctl/pctl.cpp
     hle/service/pctl/pctl.h
+    hle/service/pcv/pcv.cpp
+    hle/service/pcv/pcv.h
+    hle/service/pm/pm.cpp
+    hle/service/pm/pm.h
     hle/service/prepo/prepo.cpp
     hle/service/prepo/prepo.h
+    hle/service/psc/psc.cpp
+    hle/service/psc/psc.h
     hle/service/service.cpp
     hle/service/service.h
     hle/service/set/set.cpp
@@ -219,6 +311,8 @@ add_library(core STATIC
     hle/service/sm/sm.h
     hle/service/sockets/bsd.cpp
     hle/service/sockets/bsd.h
+    hle/service/sockets/ethc.cpp
+    hle/service/sockets/ethc.h
     hle/service/sockets/nsd.cpp
     hle/service/sockets/nsd.h
     hle/service/sockets/sfdnsres.cpp
@@ -233,12 +327,12 @@ add_library(core STATIC
     hle/service/spl/spl.h
     hle/service/ssl/ssl.cpp
     hle/service/ssl/ssl.h
+    hle/service/time/interface.cpp
+    hle/service/time/interface.h
     hle/service/time/time.cpp
     hle/service/time/time.h
-    hle/service/time/time_s.cpp
-    hle/service/time/time_s.h
-    hle/service/time/time_u.cpp
-    hle/service/time/time_u.h
+    hle/service/usb/usb.cpp
+    hle/service/usb/usb.h
     hle/service/vi/vi.cpp
     hle/service/vi/vi.h
     hle/service/vi/vi_m.cpp
@@ -247,12 +341,8 @@ add_library(core STATIC
     hle/service/vi/vi_s.h
     hle/service/vi/vi_u.cpp
     hle/service/vi/vi_u.h
-    hle/shared_page.cpp
-    hle/shared_page.h
-    hw/hw.cpp
-    hw/hw.h
-    hw/lcd.cpp
-    hw/lcd.h
+    hle/service/wlan/wlan.cpp
+    hle/service/wlan/wlan.h
     loader/deconstructed_rom_directory.cpp
     loader/deconstructed_rom_directory.h
     loader/elf.cpp
@@ -261,12 +351,16 @@ add_library(core STATIC
     loader/linker.h
     loader/loader.cpp
     loader/loader.h
+    loader/nax.cpp
+    loader/nax.h
     loader/nca.cpp
     loader/nca.h
     loader/nro.cpp
     loader/nro.h
     loader/nso.cpp
     loader/nso.h
+    loader/xci.cpp
+    loader/xci.h
     memory.cpp
     memory.h
     memory_hook.cpp
@@ -285,8 +379,8 @@ add_library(core STATIC
 
 create_target_directory_groups(core)
 
-target_link_libraries(core PUBLIC common PRIVATE video_core)
-target_link_libraries(core PUBLIC Boost::boost PRIVATE fmt lz4_static unicorn)
+target_link_libraries(core PUBLIC common PRIVATE audio_core video_core)
+target_link_libraries(core PUBLIC Boost::boost PRIVATE fmt lz4_static mbedtls opus unicorn)
 
 if (ARCHITECTURE_x86_64)
     target_sources(core PRIVATE

src/core/arm/arm_interface.h

@@ -8,6 +8,8 @@
 #include "common/common_types.h"
 #include "core/hle/kernel/vm_manager.h"
 
+namespace Core {
+
 /// Generic ARM11 CPU interface
 class ARM_Interface : NonCopyable {
 public:
@@ -20,9 +22,6 @@ public:
         u64 cpsr;
         std::array<u128, 32> fpu_registers;
         u64 fpscr;
-
-        // TODO(bunnei): Fix once we have proper support for tpidrro_el0, etc. in the JIT
-        VAddr tls_address;
     };
 
     /// Runs the CPU until an event happens
@@ -104,6 +103,10 @@ public:
 
     virtual void SetTlsAddress(VAddr address) = 0;
 
+    virtual u64 GetTPIDR_EL0() const = 0;
+
+    virtual void SetTPIDR_EL0(u64 value) = 0;
+
     /**
      * Saves the current CPU context
      * @param ctx Thread context to save
@@ -121,3 +124,5 @@ public:
     /// Prepare core for thread reschedule (if needed to correctly handle state)
     virtual void PrepareReschedule() = 0;
 };
+
+} // namespace Core

src/core/arm/dynarmic/arm_dynarmic.cpp

@@ -9,11 +9,14 @@
 #include "common/logging/log.h"
 #include "core/arm/dynarmic/arm_dynarmic.h"
 #include "core/core.h"
+#include "core/core_cpu.h"
 #include "core/core_timing.h"
-#include "core/hle/kernel/memory.h"
+#include "core/hle/kernel/process.h"
 #include "core/hle/kernel/svc.h"
 #include "core/memory.h"
 
+namespace Core {
+
 using Vector = Dynarmic::A64::Vector;
 
 class ARM_Dynarmic_Callbacks : public Dynarmic::A64::UserCallbacks {
@@ -86,7 +89,16 @@ public:
     }
 
     void AddTicks(u64 ticks) override {
-        CoreTiming::AddTicks(ticks - num_interpreted_instructions);
+        // Divide the number of ticks by the amount of CPU cores. TODO(Subv): This yields only a
+        // rough approximation of the amount of executed ticks in the system, it may be thrown off
+        // if not all cores are doing a similar amount of work. Instead of doing this, we should
+        // device a way so that timing is consistent across all cores without increasing the ticks 4
+        // times.
+        u64 amortized_ticks = (ticks - num_interpreted_instructions) / Core::NUM_CPU_CORES;
+        // Always execute at least one tick.
+        amortized_ticks = std::max<u64>(amortized_ticks, 1);
+
+        CoreTiming::AddTicks(amortized_ticks);
         num_interpreted_instructions = 0;
     }
     u64 GetTicksRemaining() override {
@@ -102,18 +114,31 @@ public:
     u64 tpidr_el0 = 0;
 };
 
-std::unique_ptr<Dynarmic::A64::Jit> MakeJit(const std::unique_ptr<ARM_Dynarmic_Callbacks>& cb) {
-    const auto page_table = Core::CurrentProcess()->vm_manager.page_table.pointers.data();
+std::unique_ptr<Dynarmic::A64::Jit> ARM_Dynarmic::MakeJit() const {
+    auto** const page_table = Core::CurrentProcess()->vm_manager.page_table.pointers.data();
 
     Dynarmic::A64::UserConfig config;
+
+    // Callbacks
     config.callbacks = cb.get();
+
+    // Memory
+    config.page_table = reinterpret_cast<void**>(page_table);
+    config.page_table_address_space_bits = Memory::ADDRESS_SPACE_BITS;
+    config.silently_mirror_page_table = false;
+
+    // Multi-process state
+    config.processor_id = core_index;
+    config.global_monitor = &exclusive_monitor->monitor;
+
+    // System registers
     config.tpidrro_el0 = &cb->tpidrro_el0;
     config.tpidr_el0 = &cb->tpidr_el0;
     config.dczid_el0 = 4;
     config.ctr_el0 = 0x8444c004;
-    config.page_table = reinterpret_cast<void**>(page_table);
-    config.page_table_address_space_bits = Memory::ADDRESS_SPACE_BITS;
-    config.silently_mirror_page_table = false;
+
+    // Unpredictable instructions
+    config.define_unpredictable_behaviour = true;
 
     return std::make_unique<Dynarmic::A64::Jit>(config);
 }
@@ -128,12 +153,13 @@ void ARM_Dynarmic::Step() {
     cb->InterpreterFallback(jit->GetPC(), 1);
 }
 
-ARM_Dynarmic::ARM_Dynarmic()
-    : cb(std::make_unique<ARM_Dynarmic_Callbacks>(*this)), jit(MakeJit(cb)) {
-    ARM_Interface::ThreadContext ctx;
+ARM_Dynarmic::ARM_Dynarmic(std::shared_ptr<ExclusiveMonitor> exclusive_monitor, size_t core_index)
+    : cb(std::make_unique<ARM_Dynarmic_Callbacks>(*this)), core_index{core_index},
+      exclusive_monitor{std::dynamic_pointer_cast<DynarmicExclusiveMonitor>(exclusive_monitor)} {
+    ThreadContext ctx;
     inner_unicorn.SaveContext(ctx);
-    LoadContext(ctx);
     PageTableChanged();
+    LoadContext(ctx);
 }
 
 ARM_Dynarmic::~ARM_Dynarmic() = default;
@@ -192,34 +218,38 @@ u64 ARM_Dynarmic::GetTlsAddress() const {
     return cb->tpidrro_el0;
 }
 
-void ARM_Dynarmic::SetTlsAddress(u64 address) {
+void ARM_Dynarmic::SetTlsAddress(VAddr address) {
     cb->tpidrro_el0 = address;
 }
 
-void ARM_Dynarmic::SaveContext(ARM_Interface::ThreadContext& ctx) {
+u64 ARM_Dynarmic::GetTPIDR_EL0() const {
+    return cb->tpidr_el0;
+}
+
+void ARM_Dynarmic::SetTPIDR_EL0(u64 value) {
+    cb->tpidr_el0 = value;
+}
+
+void ARM_Dynarmic::SaveContext(ThreadContext& ctx) {
     ctx.cpu_registers = jit->GetRegisters();
     ctx.sp = jit->GetSP();
     ctx.pc = jit->GetPC();
     ctx.cpsr = jit->GetPstate();
     ctx.fpu_registers = jit->GetVectors();
     ctx.fpscr = jit->GetFpcr();
-    ctx.tls_address = cb->tpidrro_el0;
 }
 
-void ARM_Dynarmic::LoadContext(const ARM_Interface::ThreadContext& ctx) {
+void ARM_Dynarmic::LoadContext(const ThreadContext& ctx) {
     jit->SetRegisters(ctx.cpu_registers);
     jit->SetSP(ctx.sp);
     jit->SetPC(ctx.pc);
     jit->SetPstate(static_cast<u32>(ctx.cpsr));
     jit->SetVectors(ctx.fpu_registers);
     jit->SetFpcr(static_cast<u32>(ctx.fpscr));
-    cb->tpidrro_el0 = ctx.tls_address;
 }
 
 void ARM_Dynarmic::PrepareReschedule() {
-    if (jit->IsExecuting()) {
-        jit->HaltExecution();
-    }
+    jit->HaltExecution();
 }
 
 void ARM_Dynarmic::ClearInstructionCache() {
@@ -231,6 +261,47 @@ void ARM_Dynarmic::ClearExclusiveState() {
 }
 
 void ARM_Dynarmic::PageTableChanged() {
-    jit = MakeJit(cb);
+    jit = MakeJit();
     current_page_table = Memory::GetCurrentPageTable();
 }
+
+DynarmicExclusiveMonitor::DynarmicExclusiveMonitor(size_t core_count) : monitor(core_count) {}
+DynarmicExclusiveMonitor::~DynarmicExclusiveMonitor() = default;
+
+void DynarmicExclusiveMonitor::SetExclusive(size_t core_index, VAddr addr) {
+    // Size doesn't actually matter.
+    monitor.Mark(core_index, addr, 16);
+}
+
+void DynarmicExclusiveMonitor::ClearExclusive() {
+    monitor.Clear();
+}
+
+bool DynarmicExclusiveMonitor::ExclusiveWrite8(size_t core_index, VAddr vaddr, u8 value) {
+    return monitor.DoExclusiveOperation(core_index, vaddr, 1,
+                                        [&] { Memory::Write8(vaddr, value); });
+}
+
+bool DynarmicExclusiveMonitor::ExclusiveWrite16(size_t core_index, VAddr vaddr, u16 value) {
+    return monitor.DoExclusiveOperation(core_index, vaddr, 2,
+                                        [&] { Memory::Write16(vaddr, value); });
+}
+
+bool DynarmicExclusiveMonitor::ExclusiveWrite32(size_t core_index, VAddr vaddr, u32 value) {
+    return monitor.DoExclusiveOperation(core_index, vaddr, 4,
+                                        [&] { Memory::Write32(vaddr, value); });
+}
+
+bool DynarmicExclusiveMonitor::ExclusiveWrite64(size_t core_index, VAddr vaddr, u64 value) {
+    return monitor.DoExclusiveOperation(core_index, vaddr, 8,
+                                        [&] { Memory::Write64(vaddr, value); });
+}
+
+bool DynarmicExclusiveMonitor::ExclusiveWrite128(size_t core_index, VAddr vaddr, u128 value) {
+    return monitor.DoExclusiveOperation(core_index, vaddr, 16, [&] {
+        Memory::Write64(vaddr, value[0]);
+        Memory::Write64(vaddr, value[1]);
+    });
+}
+
+} // namespace Core

src/core/arm/dynarmic/arm_dynarmic.h

@@ -6,15 +6,20 @@
 
 #include <memory>
 #include <dynarmic/A64/a64.h>
+#include <dynarmic/A64/exclusive_monitor.h>
 #include "common/common_types.h"
 #include "core/arm/arm_interface.h"
+#include "core/arm/exclusive_monitor.h"
 #include "core/arm/unicorn/arm_unicorn.h"
 
+namespace Core {
+
 class ARM_Dynarmic_Callbacks;
+class DynarmicExclusiveMonitor;
 
 class ARM_Dynarmic final : public ARM_Interface {
 public:
-    ARM_Dynarmic();
+    ARM_Dynarmic(std::shared_ptr<ExclusiveMonitor> exclusive_monitor, size_t core_index);
     ~ARM_Dynarmic();
 
     void MapBackingMemory(VAddr address, size_t size, u8* memory,
@@ -34,6 +39,8 @@ public:
     void SetCPSR(u32 cpsr) override;
     VAddr GetTlsAddress() const override;
     void SetTlsAddress(VAddr address) override;
+    void SetTPIDR_EL0(u64 value) override;
+    u64 GetTPIDR_EL0() const override;
 
     void SaveContext(ThreadContext& ctx) override;
     void LoadContext(const ThreadContext& ctx) override;
@@ -45,10 +52,36 @@ public:
     void PageTableChanged() override;
 
 private:
+    std::unique_ptr<Dynarmic::A64::Jit> MakeJit() const;
+
     friend class ARM_Dynarmic_Callbacks;
     std::unique_ptr<ARM_Dynarmic_Callbacks> cb;
     std::unique_ptr<Dynarmic::A64::Jit> jit;
     ARM_Unicorn inner_unicorn;
 
+    size_t core_index;
+    std::shared_ptr<DynarmicExclusiveMonitor> exclusive_monitor;
+
     Memory::PageTable* current_page_table = nullptr;
 };
+
+class DynarmicExclusiveMonitor final : public ExclusiveMonitor {
+public:
+    explicit DynarmicExclusiveMonitor(size_t core_count);
+    ~DynarmicExclusiveMonitor();
+
+    void SetExclusive(size_t core_index, VAddr addr) override;
+    void ClearExclusive() override;
+
+    bool ExclusiveWrite8(size_t core_index, VAddr vaddr, u8 value) override;
+    bool ExclusiveWrite16(size_t core_index, VAddr vaddr, u16 value) override;
+    bool ExclusiveWrite32(size_t core_index, VAddr vaddr, u32 value) override;
+    bool ExclusiveWrite64(size_t core_index, VAddr vaddr, u64 value) override;
+    bool ExclusiveWrite128(size_t core_index, VAddr vaddr, u128 value) override;
+
+private:
+    friend class ARM_Dynarmic;
+    Dynarmic::A64::ExclusiveMonitor monitor;
+};
+
+} // namespace Core

src/core/arm/exclusive_monitor.cpp

@@ -0,0 +1,11 @@
+// Copyright 2018 yuzu emulator team
+// Licensed under GPLv2 or any later version
+// Refer to the license.txt file included.
+
+#include "core/arm/exclusive_monitor.h"
+
+namespace Core {
+
+ExclusiveMonitor::~ExclusiveMonitor() = default;
+
+} // namespace Core

src/core/arm/exclusive_monitor.h

@@ -0,0 +1,25 @@
+// Copyright 2018 yuzu emulator team
+// Licensed under GPLv2 or any later version
+// Refer to the license.txt file included.
+
+#pragma once
+
+#include "common/common_types.h"
+
+namespace Core {
+
+class ExclusiveMonitor {
+public:
+    virtual ~ExclusiveMonitor();
+
+    virtual void SetExclusive(size_t core_index, VAddr addr) = 0;
+    virtual void ClearExclusive() = 0;
+
+    virtual bool ExclusiveWrite8(size_t core_index, VAddr vaddr, u8 value) = 0;
+    virtual bool ExclusiveWrite16(size_t core_index, VAddr vaddr, u16 value) = 0;
+    virtual bool ExclusiveWrite32(size_t core_index, VAddr vaddr, u32 value) = 0;
+    virtual bool ExclusiveWrite64(size_t core_index, VAddr vaddr, u64 value) = 0;
+    virtual bool ExclusiveWrite128(size_t core_index, VAddr vaddr, u128 value) = 0;
+};
+
+} // namespace Core

src/core/arm/unicorn/arm_unicorn.cpp

@@ -11,6 +11,8 @@
 #include "core/core_timing.h"
 #include "core/hle/kernel/svc.h"
 
+namespace Core {
+
 // Load Unicorn DLL once on Windows using RAII
 #ifdef _MSC_VER
 #include <unicorn_dynload.h>
@@ -169,6 +171,16 @@ void ARM_Unicorn::SetTlsAddress(VAddr base) {
     CHECKED(uc_reg_write(uc, UC_ARM64_REG_TPIDRRO_EL0, &base));
 }
 
+u64 ARM_Unicorn::GetTPIDR_EL0() const {
+    u64 value{};
+    CHECKED(uc_reg_read(uc, UC_ARM64_REG_TPIDR_EL0, &value));
+    return value;
+}
+
+void ARM_Unicorn::SetTPIDR_EL0(u64 value) {
+    CHECKED(uc_reg_write(uc, UC_ARM64_REG_TPIDR_EL0, &value));
+}
+
 void ARM_Unicorn::Run() {
     if (GDBStub::IsServerEnabled()) {
         ExecuteInstructions(std::max(4000000, 0));
@@ -193,7 +205,7 @@ void ARM_Unicorn::ExecuteInstructions(int num_instructions) {
         }
         Kernel::Thread* thread = Kernel::GetCurrentThread();
         SaveContext(thread->context);
-        if (last_bkpt_hit || (num_instructions == 1)) {
+        if (last_bkpt_hit || GDBStub::GetCpuStepFlag()) {
             last_bkpt_hit = false;
             GDBStub::Break();
             GDBStub::SendTrap(thread, 5);
@@ -201,7 +213,7 @@ void ARM_Unicorn::ExecuteInstructions(int num_instructions) {
     }
 }
 
-void ARM_Unicorn::SaveContext(ARM_Interface::ThreadContext& ctx) {
+void ARM_Unicorn::SaveContext(ThreadContext& ctx) {
     int uregs[32];
     void* tregs[32];
 
@@ -220,8 +232,6 @@ void ARM_Unicorn::SaveContext(ARM_Interface::ThreadContext& ctx) {
 
     CHECKED(uc_reg_read_batch(uc, uregs, tregs, 31));
 
-    ctx.tls_address = GetTlsAddress();
-
     for (int i = 0; i < 32; ++i) {
         uregs[i] = UC_ARM64_REG_Q0 + i;
         tregs[i] = &ctx.fpu_registers[i];
@@ -230,7 +240,7 @@ void ARM_Unicorn::SaveContext(ARM_Interface::ThreadContext& ctx) {
     CHECKED(uc_reg_read_batch(uc, uregs, tregs, 32));
 }
 
-void ARM_Unicorn::LoadContext(const ARM_Interface::ThreadContext& ctx) {
+void ARM_Unicorn::LoadContext(const ThreadContext& ctx) {
     int uregs[32];
     void* tregs[32];
 
@@ -249,8 +259,6 @@ void ARM_Unicorn::LoadContext(const ARM_Interface::ThreadContext& ctx) {
 
     CHECKED(uc_reg_write_batch(uc, uregs, tregs, 31));
 
-    SetTlsAddress(ctx.tls_address);
-
     for (auto i = 0; i < 32; ++i) {
         uregs[i] = UC_ARM64_REG_Q0 + i;
         tregs[i] = (void*)&ctx.fpu_registers[i];
@@ -271,3 +279,5 @@ void ARM_Unicorn::RecordBreak(GDBStub::BreakpointAddress bkpt) {
     last_bkpt = bkpt;
     last_bkpt_hit = true;
 }
+
+} // namespace Core

src/core/arm/unicorn/arm_unicorn.h

@@ -9,6 +9,8 @@
 #include "core/arm/arm_interface.h"
 #include "core/gdbstub/gdbstub.h"
 
+namespace Core {
+
 class ARM_Unicorn final : public ARM_Interface {
 public:
     ARM_Unicorn();
@@ -28,6 +30,8 @@ public:
     void SetCPSR(u32 cpsr) override;
     VAddr GetTlsAddress() const override;
     void SetTlsAddress(VAddr address) override;
+    void SetTPIDR_EL0(u64 value) override;
+    u64 GetTPIDR_EL0() const override;
     void SaveContext(ThreadContext& ctx) override;
     void LoadContext(const ThreadContext& ctx) override;
     void PrepareReschedule() override;
@@ -44,3 +48,5 @@ private:
     GDBStub::BreakpointAddress last_bkpt{};
     bool last_bkpt_hit;
 };
+
+} // namespace Core

src/core/core.cpp

@@ -2,261 +2,471 @@
 // Licensed under GPLv2 or any later version
 // Refer to the license.txt file included.
 
+#include <array>
+#include <map>
 #include <memory>
+#include <thread>
 #include <utility>
+
 #include "common/logging/log.h"
+#include "common/string_util.h"
+#include "core/arm/exclusive_monitor.h"
 #include "core/core.h"
+#include "core/core_cpu.h"
 #include "core/core_timing.h"
+#include "core/file_sys/mode.h"
+#include "core/file_sys/vfs_concat.h"
+#include "core/file_sys/vfs_real.h"
 #include "core/gdbstub/gdbstub.h"
 #include "core/hle/kernel/client_port.h"
 #include "core/hle/kernel/kernel.h"
 #include "core/hle/kernel/process.h"
+#include "core/hle/kernel/scheduler.h"
 #include "core/hle/kernel/thread.h"
 #include "core/hle/service/service.h"
 #include "core/hle/service/sm/controller.h"
 #include "core/hle/service/sm/sm.h"
-#include "core/hw/hw.h"
 #include "core/loader/loader.h"
-#include "core/memory_setup.h"
+#include "core/perf_stats.h"
 #include "core/settings.h"
-#include "file_sys/vfs_real.h"
+#include "core/telemetry_session.h"
+#include "video_core/debug_utils/debug_utils.h"
+#include "video_core/gpu.h"
+#include "video_core/renderer_base.h"
 #include "video_core/video_core.h"
 
 namespace Core {
 
 /*static*/ System System::s_instance;
 
-System::System() = default;
+namespace {
+FileSys::VirtualFile GetGameFileFromPath(const FileSys::VirtualFilesystem& vfs,
+                                         const std::string& path) {
+    // To account for split 00+01+etc files.
+    std::string dir_name;
+    std::string filename;
+    Common::SplitPath(path, &dir_name, &filename, nullptr);
+    if (filename == "00") {
+        const auto dir = vfs->OpenDirectory(dir_name, FileSys::Mode::Read);
+        std::vector<FileSys::VirtualFile> concat;
+        for (u8 i = 0; i < 0x10; ++i) {
+            auto next = dir->GetFile(fmt::format("{:02X}", i));
+            if (next != nullptr)
+                concat.push_back(std::move(next));
+            else {
+                next = dir->GetFile(fmt::format("{:02x}", i));
+                if (next != nullptr)
+                    concat.push_back(std::move(next));
+                else
+                    break;
+            }
+        }
 
-System::~System() = default;
+        if (concat.empty())
+            return nullptr;
+
+        return FileSys::ConcatenateFiles(concat, dir->GetName());
+    }
+
+    return vfs->OpenFile(path, FileSys::Mode::Read);
+}
 
 /// Runs a CPU core while the system is powered on
-static void RunCpuCore(std::shared_ptr<Cpu> cpu_state) {
+void RunCpuCore(std::shared_ptr<Cpu> cpu_state) {
     while (Core::System::GetInstance().IsPoweredOn()) {
         cpu_state->RunLoop(true);
     }
 }
+} // Anonymous namespace
 
-Cpu& System::CurrentCpuCore() {
-    // If multicore is enabled, use host thread to figure out the current CPU core
-    if (Settings::values.use_multi_core) {
-        const auto& search = thread_to_cpu.find(std::this_thread::get_id());
-        ASSERT(search != thread_to_cpu.end());
-        ASSERT(search->second);
-        return *search->second;
+struct System::Impl {
+    Cpu& CurrentCpuCore() {
+        if (Settings::values.use_multi_core) {
+            const auto& search = thread_to_cpu.find(std::this_thread::get_id());
+            ASSERT(search != thread_to_cpu.end());
+            ASSERT(search->second);
+            return *search->second;
+        }
+
+        // Otherwise, use single-threaded mode active_core variable
+        return *cpu_cores[active_core];
     }
 
-    // Otherwise, use single-threaded mode active_core variable
-    return *cpu_cores[active_core];
+    ResultStatus RunLoop(bool tight_loop) {
+        status = ResultStatus::Success;
+
+        // Update thread_to_cpu in case Core 0 is run from a different host thread
+        thread_to_cpu[std::this_thread::get_id()] = cpu_cores[0];
+
+        if (GDBStub::IsServerEnabled()) {
+            GDBStub::HandlePacket();
+
+            // If the loop is halted and we want to step, use a tiny (1) number of instructions to
+            // execute. Otherwise, get out of the loop function.
+            if (GDBStub::GetCpuHaltFlag()) {
+                if (GDBStub::GetCpuStepFlag()) {
+                    tight_loop = false;
+                } else {
+                    return ResultStatus::Success;
+                }
+            }
+        }
+
+        for (active_core = 0; active_core < NUM_CPU_CORES; ++active_core) {
+            cpu_cores[active_core]->RunLoop(tight_loop);
+            if (Settings::values.use_multi_core) {
+                // Cores 1-3 are run on other threads in this mode
+                break;
+            }
+        }
+
+        if (GDBStub::IsServerEnabled()) {
+            GDBStub::SetCpuStepFlag(false);
+        }
+
+        return status;
+    }
+
+    ResultStatus Init(Frontend::EmuWindow& emu_window) {
+        LOG_DEBUG(HW_Memory, "initialized OK");
+
+        CoreTiming::Init();
+        kernel.Initialize();
+
+        // Create a default fs if one doesn't already exist.
+        if (virtual_filesystem == nullptr)
+            virtual_filesystem = std::make_shared<FileSys::RealVfsFilesystem>();
+
+        current_process = Kernel::Process::Create(kernel, "main");
+
+        cpu_barrier = std::make_shared<CpuBarrier>();
+        cpu_exclusive_monitor = Cpu::MakeExclusiveMonitor(cpu_cores.size());
+        for (size_t index = 0; index < cpu_cores.size(); ++index) {
+            cpu_cores[index] = std::make_shared<Cpu>(cpu_exclusive_monitor, cpu_barrier, index);
+        }
+
+        telemetry_session = std::make_unique<Core::TelemetrySession>();
+        service_manager = std::make_shared<Service::SM::ServiceManager>();
+
+        Service::Init(service_manager, virtual_filesystem);
+        GDBStub::Init();
+
+        renderer = VideoCore::CreateRenderer(emu_window);
+        if (!renderer->Init()) {
+            return ResultStatus::ErrorVideoCore;
+        }
+
+        gpu_core = std::make_unique<Tegra::GPU>(renderer->Rasterizer());
+
+        // Create threads for CPU cores 1-3, and build thread_to_cpu map
+        // CPU core 0 is run on the main thread
+        thread_to_cpu[std::this_thread::get_id()] = cpu_cores[0];
+        if (Settings::values.use_multi_core) {
+            for (size_t index = 0; index < cpu_core_threads.size(); ++index) {
+                cpu_core_threads[index] =
+                    std::make_unique<std::thread>(RunCpuCore, cpu_cores[index + 1]);
+                thread_to_cpu[cpu_core_threads[index]->get_id()] = cpu_cores[index + 1];
+            }
+        }
+
+        LOG_DEBUG(Core, "Initialized OK");
+
+        // Reset counters and set time origin to current frame
+        GetAndResetPerfStats();
+        perf_stats.BeginSystemFrame();
+
+        return ResultStatus::Success;
+    }
+
+    ResultStatus Load(Frontend::EmuWindow& emu_window, const std::string& filepath) {
+        app_loader = Loader::GetLoader(GetGameFileFromPath(virtual_filesystem, filepath));
+
+        if (!app_loader) {
+            LOG_CRITICAL(Core, "Failed to obtain loader for {}!", filepath);
+            return ResultStatus::ErrorGetLoader;
+        }
+        std::pair<boost::optional<u32>, Loader::ResultStatus> system_mode =
+            app_loader->LoadKernelSystemMode();
+
+        if (system_mode.second != Loader::ResultStatus::Success) {
+            LOG_CRITICAL(Core, "Failed to determine system mode (Error {})!",
+                         static_cast<int>(system_mode.second));
+
+            return ResultStatus::ErrorSystemMode;
+        }
+
+        ResultStatus init_result{Init(emu_window)};
+        if (init_result != ResultStatus::Success) {
+            LOG_CRITICAL(Core, "Failed to initialize system (Error {})!",
+                         static_cast<int>(init_result));
+            Shutdown();
+            return init_result;
+        }
+
+        const Loader::ResultStatus load_result{app_loader->Load(current_process)};
+        if (load_result != Loader::ResultStatus::Success) {
+            LOG_CRITICAL(Core, "Failed to load ROM (Error {})!", static_cast<int>(load_result));
+            Shutdown();
+
+            return static_cast<ResultStatus>(static_cast<u32>(ResultStatus::ErrorLoader) +
+                                             static_cast<u32>(load_result));
+        }
+        status = ResultStatus::Success;
+        return status;
+    }
+
+    void Shutdown() {
+        // Log last frame performance stats
+        auto perf_results = GetAndResetPerfStats();
+        Telemetry().AddField(Telemetry::FieldType::Performance, "Shutdown_EmulationSpeed",
+                             perf_results.emulation_speed * 100.0);
+        Telemetry().AddField(Telemetry::FieldType::Performance, "Shutdown_Framerate",
+                             perf_results.game_fps);
+        Telemetry().AddField(Telemetry::FieldType::Performance, "Shutdown_Frametime",
+                             perf_results.frametime * 1000.0);
+
+        // Shutdown emulation session
+        renderer.reset();
+        GDBStub::Shutdown();
+        Service::Shutdown();
+        service_manager.reset();
+        telemetry_session.reset();
+        gpu_core.reset();
+
+        // Close all CPU/threading state
+        cpu_barrier->NotifyEnd();
+        if (Settings::values.use_multi_core) {
+            for (auto& thread : cpu_core_threads) {
+                thread->join();
+                thread.reset();
+            }
+        }
+        thread_to_cpu.clear();
+        for (auto& cpu_core : cpu_cores) {
+            cpu_core.reset();
+        }
+        cpu_barrier.reset();
+
+        // Shutdown kernel and core timing
+        kernel.Shutdown();
+        CoreTiming::Shutdown();
+
+        // Close app loader
+        app_loader.reset();
+
+        LOG_DEBUG(Core, "Shutdown OK");
+    }
+
+    Loader::ResultStatus GetGameName(std::string& out) const {
+        if (app_loader == nullptr)
+            return Loader::ResultStatus::ErrorNotInitialized;
+        return app_loader->ReadTitle(out);
+    }
+
+    void SetStatus(ResultStatus new_status, const char* details = nullptr) {
+        status = new_status;
+        if (details) {
+            status_details = details;
+        }
+    }
+
+    PerfStatsResults GetAndResetPerfStats() {
+        return perf_stats.GetAndResetStats(CoreTiming::GetGlobalTimeUs());
+    }
+
+    Kernel::KernelCore kernel;
+    /// RealVfsFilesystem instance
+    FileSys::VirtualFilesystem virtual_filesystem;
+    /// AppLoader used to load the current executing application
+    std::unique_ptr<Loader::AppLoader> app_loader;
+    std::unique_ptr<VideoCore::RendererBase> renderer;
+    std::unique_ptr<Tegra::GPU> gpu_core;
+    std::shared_ptr<Tegra::DebugContext> debug_context;
+    Kernel::SharedPtr<Kernel::Process> current_process;
+    std::shared_ptr<ExclusiveMonitor> cpu_exclusive_monitor;
+    std::shared_ptr<CpuBarrier> cpu_barrier;
+    std::array<std::shared_ptr<Cpu>, NUM_CPU_CORES> cpu_cores;
+    std::array<std::unique_ptr<std::thread>, NUM_CPU_CORES - 1> cpu_core_threads;
+    size_t active_core{}; ///< Active core, only used in single thread mode
+
+    /// Service manager
+    std::shared_ptr<Service::SM::ServiceManager> service_manager;
+
+    /// Telemetry session for this emulation session
+    std::unique_ptr<Core::TelemetrySession> telemetry_session;
+
+    ResultStatus status = ResultStatus::Success;
+    std::string status_details = "";
+
+    /// Map of guest threads to CPU cores
+    std::map<std::thread::id, std::shared_ptr<Cpu>> thread_to_cpu;
+
+    Core::PerfStats perf_stats;
+    Core::FrameLimiter frame_limiter;
+};
+
+System::System() : impl{std::make_unique<Impl>()} {}
+System::~System() = default;
+
+Cpu& System::CurrentCpuCore() {
+    return impl->CurrentCpuCore();
 }
 
 System::ResultStatus System::RunLoop(bool tight_loop) {
-    status = ResultStatus::Success;
-
-    // Update thread_to_cpu in case Core 0 is run from a different host thread
-    thread_to_cpu[std::this_thread::get_id()] = cpu_cores[0];
-
-    if (GDBStub::IsServerEnabled()) {
-        GDBStub::HandlePacket();
-
-        // If the loop is halted and we want to step, use a tiny (1) number of instructions to
-        // execute. Otherwise, get out of the loop function.
-        if (GDBStub::GetCpuHaltFlag()) {
-            if (GDBStub::GetCpuStepFlag()) {
-                GDBStub::SetCpuStepFlag(false);
-                tight_loop = false;
-            } else {
-                return ResultStatus::Success;
-            }
-        }
-    }
-
-    for (active_core = 0; active_core < NUM_CPU_CORES; ++active_core) {
-        cpu_cores[active_core]->RunLoop(tight_loop);
-        if (Settings::values.use_multi_core) {
-            // Cores 1-3 are run on other threads in this mode
-            break;
-        }
-    }
-
-    return status;
+    return impl->RunLoop(tight_loop);
 }
 
 System::ResultStatus System::SingleStep() {
     return RunLoop(false);
 }
 
-System::ResultStatus System::Load(EmuWindow* emu_window, const std::string& filepath) {
-    app_loader = Loader::GetLoader(std::make_shared<FileSys::RealVfsFile>(filepath));
-
-    if (!app_loader) {
-        LOG_CRITICAL(Core, "Failed to obtain loader for {}!", filepath);
-        return ResultStatus::ErrorGetLoader;
+void System::InvalidateCpuInstructionCaches() {
+    for (auto& cpu : impl->cpu_cores) {
+        cpu->ArmInterface().ClearInstructionCache();
     }
-    std::pair<boost::optional<u32>, Loader::ResultStatus> system_mode =
-        app_loader->LoadKernelSystemMode();
+}
 
-    if (system_mode.second != Loader::ResultStatus::Success) {
-        LOG_CRITICAL(Core, "Failed to determine system mode (Error {})!",
-                     static_cast<int>(system_mode.second));
+System::ResultStatus System::Load(Frontend::EmuWindow& emu_window, const std::string& filepath) {
+    return impl->Load(emu_window, filepath);
+}
 
-        switch (system_mode.second) {
-        case Loader::ResultStatus::ErrorEncrypted:
-            return ResultStatus::ErrorLoader_ErrorEncrypted;
-        case Loader::ResultStatus::ErrorInvalidFormat:
-            return ResultStatus::ErrorLoader_ErrorInvalidFormat;
-        case Loader::ResultStatus::ErrorUnsupportedArch:
-            return ResultStatus::ErrorUnsupportedArch;
-        default:
-            return ResultStatus::ErrorSystemMode;
-        }
-    }
-
-    ResultStatus init_result{Init(emu_window, system_mode.first.get())};
-    if (init_result != ResultStatus::Success) {
-        LOG_CRITICAL(Core, "Failed to initialize system (Error {})!",
-                     static_cast<int>(init_result));
-        System::Shutdown();
-        return init_result;
-    }
-
-    const Loader::ResultStatus load_result{app_loader->Load(current_process)};
-    if (Loader::ResultStatus::Success != load_result) {
-        LOG_CRITICAL(Core, "Failed to load ROM (Error {})!", static_cast<int>(load_result));
-        System::Shutdown();
-
-        switch (load_result) {
-        case Loader::ResultStatus::ErrorEncrypted:
-            return ResultStatus::ErrorLoader_ErrorEncrypted;
-        case Loader::ResultStatus::ErrorInvalidFormat:
-            return ResultStatus::ErrorLoader_ErrorInvalidFormat;
-        case Loader::ResultStatus::ErrorUnsupportedArch:
-            return ResultStatus::ErrorUnsupportedArch;
-        default:
-            return ResultStatus::ErrorLoader;
-        }
-    }
-    status = ResultStatus::Success;
-    return status;
+bool System::IsPoweredOn() const {
+    return impl->cpu_barrier && impl->cpu_barrier->IsAlive();
 }
 
 void System::PrepareReschedule() {
     CurrentCpuCore().PrepareReschedule();
 }
 
-PerfStats::Results System::GetAndResetPerfStats() {
-    return perf_stats.GetAndResetStats(CoreTiming::GetGlobalTimeUs());
+PerfStatsResults System::GetAndResetPerfStats() {
+    return impl->GetAndResetPerfStats();
+}
+
+Core::TelemetrySession& System::TelemetrySession() const {
+    return *impl->telemetry_session;
+}
+
+ARM_Interface& System::CurrentArmInterface() {
+    return CurrentCpuCore().ArmInterface();
+}
+
+size_t System::CurrentCoreIndex() {
+    return CurrentCpuCore().CoreIndex();
+}
+
+Kernel::Scheduler& System::CurrentScheduler() {
+    return *CurrentCpuCore().Scheduler();
 }
 
 const std::shared_ptr<Kernel::Scheduler>& System::Scheduler(size_t core_index) {
     ASSERT(core_index < NUM_CPU_CORES);
-    return cpu_cores[core_index]->Scheduler();
+    return impl->cpu_cores[core_index]->Scheduler();
+}
+
+Kernel::SharedPtr<Kernel::Process>& System::CurrentProcess() {
+    return impl->current_process;
 }
 
 ARM_Interface& System::ArmInterface(size_t core_index) {
     ASSERT(core_index < NUM_CPU_CORES);
-    return cpu_cores[core_index]->ArmInterface();
+    return impl->cpu_cores[core_index]->ArmInterface();
 }
 
 Cpu& System::CpuCore(size_t core_index) {
     ASSERT(core_index < NUM_CPU_CORES);
-    return *cpu_cores[core_index];
+    return *impl->cpu_cores[core_index];
 }
 
-System::ResultStatus System::Init(EmuWindow* emu_window, u32 system_mode) {
-    LOG_DEBUG(HW_Memory, "initialized OK");
+ExclusiveMonitor& System::Monitor() {
+    return *impl->cpu_exclusive_monitor;
+}
 
-    CoreTiming::Init();
+Tegra::GPU& System::GPU() {
+    return *impl->gpu_core;
+}
 
-    current_process = Kernel::Process::Create("main");
+const Tegra::GPU& System::GPU() const {
+    return *impl->gpu_core;
+}
 
-    cpu_barrier = std::make_shared<CpuBarrier>();
-    for (size_t index = 0; index < cpu_cores.size(); ++index) {
-        cpu_cores[index] = std::make_shared<Cpu>(cpu_barrier, index);
-    }
+VideoCore::RendererBase& System::Renderer() {
+    return *impl->renderer;
+}
 
-    gpu_core = std::make_unique<Tegra::GPU>();
-    telemetry_session = std::make_unique<Core::TelemetrySession>();
-    service_manager = std::make_shared<Service::SM::ServiceManager>();
+const VideoCore::RendererBase& System::Renderer() const {
+    return *impl->renderer;
+}
 
-    HW::Init();
-    Kernel::Init(system_mode);
-    Service::Init(service_manager);
-    GDBStub::Init();
+Kernel::KernelCore& System::Kernel() {
+    return impl->kernel;
+}
 
-    if (!VideoCore::Init(emu_window)) {
-        return ResultStatus::ErrorVideoCore;
-    }
+const Kernel::KernelCore& System::Kernel() const {
+    return impl->kernel;
+}
 
-    // Create threads for CPU cores 1-3, and build thread_to_cpu map
-    // CPU core 0 is run on the main thread
-    thread_to_cpu[std::this_thread::get_id()] = cpu_cores[0];
-    if (Settings::values.use_multi_core) {
-        for (size_t index = 0; index < cpu_core_threads.size(); ++index) {
-            cpu_core_threads[index] =
-                std::make_unique<std::thread>(RunCpuCore, cpu_cores[index + 1]);
-            thread_to_cpu[cpu_core_threads[index]->get_id()] = cpu_cores[index + 1];
-        }
-    }
+Core::PerfStats& System::GetPerfStats() {
+    return impl->perf_stats;
+}
 
-    LOG_DEBUG(Core, "Initialized OK");
+const Core::PerfStats& System::GetPerfStats() const {
+    return impl->perf_stats;
+}
 
-    // Reset counters and set time origin to current frame
-    GetAndResetPerfStats();
-    perf_stats.BeginSystemFrame();
+Core::FrameLimiter& System::FrameLimiter() {
+    return impl->frame_limiter;
+}
 
-    return ResultStatus::Success;
+const Core::FrameLimiter& System::FrameLimiter() const {
+    return impl->frame_limiter;
+}
+
+Loader::ResultStatus System::GetGameName(std::string& out) const {
+    return impl->GetGameName(out);
+}
+
+void System::SetStatus(ResultStatus new_status, const char* details) {
+    impl->SetStatus(new_status, details);
+}
+
+const std::string& System::GetStatusDetails() const {
+    return impl->status_details;
+}
+
+Loader::AppLoader& System::GetAppLoader() const {
+    return *impl->app_loader;
+}
+
+void System::SetGPUDebugContext(std::shared_ptr<Tegra::DebugContext> context) {
+    impl->debug_context = std::move(context);
+}
+
+Tegra::DebugContext* System::GetGPUDebugContext() const {
+    return impl->debug_context.get();
+}
+
+void System::SetFilesystem(std::shared_ptr<FileSys::VfsFilesystem> vfs) {
+    impl->virtual_filesystem = std::move(vfs);
+}
+
+std::shared_ptr<FileSys::VfsFilesystem> System::GetFilesystem() const {
+    return impl->virtual_filesystem;
+}
+
+System::ResultStatus System::Init(Frontend::EmuWindow& emu_window) {
+    return impl->Init(emu_window);
 }
 
 void System::Shutdown() {
-    // Log last frame performance stats
-    auto perf_results = GetAndResetPerfStats();
-    Telemetry().AddField(Telemetry::FieldType::Performance, "Shutdown_EmulationSpeed",
-                         perf_results.emulation_speed * 100.0);
-    Telemetry().AddField(Telemetry::FieldType::Performance, "Shutdown_Framerate",
-                         perf_results.game_fps);
-    Telemetry().AddField(Telemetry::FieldType::Performance, "Shutdown_Frametime",
-                         perf_results.frametime * 1000.0);
-
-    // Shutdown emulation session
-    VideoCore::Shutdown();
-    GDBStub::Shutdown();
-    Service::Shutdown();
-    Kernel::Shutdown();
-    HW::Shutdown();
-    service_manager.reset();
-    telemetry_session.reset();
-    gpu_core.reset();
-
-    // Close all CPU/threading state
-    cpu_barrier->NotifyEnd();
-    if (Settings::values.use_multi_core) {
-        for (auto& thread : cpu_core_threads) {
-            thread->join();
-            thread.reset();
-        }
-    }
-    thread_to_cpu.clear();
-    for (auto& cpu_core : cpu_cores) {
-        cpu_core.reset();
-    }
-    cpu_barrier.reset();
-
-    // Close core timing
-    CoreTiming::Shutdown();
-
-    // Close app loader
-    app_loader.reset();
-
-    LOG_DEBUG(Core, "Shutdown OK");
+    impl->Shutdown();
 }
 
 Service::SM::ServiceManager& System::ServiceManager() {
-    return *service_manager;
+    return *impl->service_manager;
 }
 
 const Service::SM::ServiceManager& System::ServiceManager() const {
-    return *service_manager;
+    return *impl->service_manager;
 }
 
 } // namespace Core

src/core/core.h

@@ -4,32 +4,64 @@
 
 #pragma once
 
-#include <array>
+#include <cstddef>
 #include <memory>
 #include <string>
-#include <thread>
-#include "common/common_types.h"
-#include "core/core_cpu.h"
-#include "core/hle/kernel/kernel.h"
-#include "core/hle/kernel/scheduler.h"
-#include "core/loader/loader.h"
-#include "core/memory.h"
-#include "core/perf_stats.h"
-#include "core/telemetry_session.h"
-#include "video_core/debug_utils/debug_utils.h"
-#include "video_core/gpu.h"
 
+#include "common/common_types.h"
+#include "core/hle/kernel/object.h"
+
+namespace Core::Frontend {
 class EmuWindow;
-class ARM_Interface;
+} // namespace Core::Frontend
+
+namespace FileSys {
+class VfsFilesystem;
+} // namespace FileSys
+
+namespace Kernel {
+class KernelCore;
+class Process;
+class Scheduler;
+} // namespace Kernel
+
+namespace Loader {
+class AppLoader;
+enum class ResultStatus : u16;
+} // namespace Loader
 
 namespace Service::SM {
 class ServiceManager;
-}
+} // namespace Service::SM
+
+namespace Tegra {
+class DebugContext;
+class GPU;
+} // namespace Tegra
+
+namespace VideoCore {
+class RendererBase;
+} // namespace VideoCore
 
 namespace Core {
 
+class ARM_Interface;
+class Cpu;
+class ExclusiveMonitor;
+class FrameLimiter;
+class PerfStats;
+class TelemetrySession;
+
+struct PerfStatsResults;
+
 class System {
 public:
+    System(const System&) = delete;
+    System& operator=(const System&) = delete;
+
+    System(System&&) = delete;
+    System& operator=(System&&) = delete;
+
     ~System();
 
     /**
@@ -42,19 +74,15 @@ public:
 
     /// Enumeration representing the return values of the System Initialize and Load process.
     enum class ResultStatus : u32 {
-        Success,                    ///< Succeeded
-        ErrorNotInitialized,        ///< Error trying to use core prior to initialization
-        ErrorGetLoader,             ///< Error finding the correct application loader
-        ErrorSystemMode,            ///< Error determining the system mode
-        ErrorLoader,                ///< Error loading the specified application
-        ErrorLoader_ErrorEncrypted, ///< Error loading the specified application due to encryption
-        ErrorLoader_ErrorInvalidFormat, ///< Error loading the specified application due to an
-                                        /// invalid format
-        ErrorSystemFiles,               ///< Error in finding system files
-        ErrorSharedFont,                ///< Error in finding shared font
-        ErrorVideoCore,                 ///< Error in the video core
-        ErrorUnsupportedArch,           ///< Unsupported Architecture (32-Bit ROMs)
-        ErrorUnknown                    ///< Any other error
+        Success,             ///< Succeeded
+        ErrorNotInitialized, ///< Error trying to use core prior to initialization
+        ErrorGetLoader,      ///< Error finding the correct application loader
+        ErrorSystemMode,     ///< Error determining the system mode
+        ErrorSystemFiles,    ///< Error in finding system files
+        ErrorSharedFont,     ///< Error in finding shared font
+        ErrorVideoCore,      ///< Error in the video core
+        ErrorUnknown,        ///< Any other error
+        ErrorLoader,         ///< The base for loader errors (too many to repeat)
     };
 
     /**
@@ -75,44 +103,52 @@ public:
      */
     ResultStatus SingleStep();
 
+    /**
+     * Invalidate the CPU instruction caches
+     * This function should only be used by GDB Stub to support breakpoints, memory updates and
+     * step/continue commands.
+     */
+    void InvalidateCpuInstructionCaches();
+
     /// Shutdown the emulated system.
     void Shutdown();
 
     /**
      * Load an executable application.
-     * @param emu_window Pointer to the host-system window used for video output and keyboard input.
+     * @param emu_window Reference to the host-system window used for video output and keyboard
+     *                   input.
      * @param filepath String path to the executable application to load on the host file system.
      * @returns ResultStatus code, indicating if the operation succeeded.
      */
-    ResultStatus Load(EmuWindow* emu_window, const std::string& filepath);
+    ResultStatus Load(Frontend::EmuWindow& emu_window, const std::string& filepath);
 
     /**
      * Indicates if the emulated system is powered on (all subsystems initialized and able to run an
      * application).
      * @returns True if the emulated system is powered on, otherwise false.
      */
-    bool IsPoweredOn() const {
-        return cpu_barrier && cpu_barrier->IsAlive();
-    }
+    bool IsPoweredOn() const;
 
     /**
      * Returns a reference to the telemetry session for this emulation session.
      * @returns Reference to the telemetry session.
      */
-    Core::TelemetrySession& TelemetrySession() const {
-        return *telemetry_session;
-    }
+    Core::TelemetrySession& TelemetrySession() const;
 
     /// Prepare the core emulation for a reschedule
     void PrepareReschedule();
 
     /// Gets and resets core performance statistics
-    PerfStats::Results GetAndResetPerfStats();
+    PerfStatsResults GetAndResetPerfStats();
 
     /// Gets an ARM interface to the CPU core that is currently running
-    ARM_Interface& CurrentArmInterface() {
-        return CurrentCpuCore().ArmInterface();
-    }
+    ARM_Interface& CurrentArmInterface();
+
+    /// Gets the index of the currently running CPU core
+    size_t CurrentCoreIndex();
+
+    /// Gets the scheduler for the CPU core that is currently running
+    Kernel::Scheduler& CurrentScheduler();
 
     /// Gets an ARM interface to the CPU core with the specified index
     ARM_Interface& ArmInterface(size_t core_index);
@@ -120,52 +156,64 @@ public:
     /// Gets a CPU interface to the CPU core with the specified index
     Cpu& CpuCore(size_t core_index);
 
-    /// Gets the GPU interface
-    Tegra::GPU& GPU() {
-        return *gpu_core;
-    }
+    /// Gets the exclusive monitor
+    ExclusiveMonitor& Monitor();
 
-    /// Gets the scheduler for the CPU core that is currently running
-    Kernel::Scheduler& CurrentScheduler() {
-        return *CurrentCpuCore().Scheduler();
-    }
+    /// Gets a mutable reference to the GPU interface
+    Tegra::GPU& GPU();
+
+    /// Gets an immutable reference to the GPU interface.
+    const Tegra::GPU& GPU() const;
+
+    /// Gets a mutable reference to the renderer.
+    VideoCore::RendererBase& Renderer();
+
+    /// Gets an immutable reference to the renderer.
+    const VideoCore::RendererBase& Renderer() const;
 
     /// Gets the scheduler for the CPU core with the specified index
     const std::shared_ptr<Kernel::Scheduler>& Scheduler(size_t core_index);
 
     /// Gets the current process
-    Kernel::SharedPtr<Kernel::Process>& CurrentProcess() {
-        return current_process;
-    }
+    Kernel::SharedPtr<Kernel::Process>& CurrentProcess();
 
-    PerfStats perf_stats;
-    FrameLimiter frame_limiter;
+    /// Provides a reference to the kernel instance.
+    Kernel::KernelCore& Kernel();
 
-    void SetStatus(ResultStatus new_status, const char* details = nullptr) {
-        status = new_status;
-        if (details) {
-            status_details = details;
-        }
-    }
+    /// Provides a constant reference to the kernel instance.
+    const Kernel::KernelCore& Kernel() const;
 
-    const std::string& GetStatusDetails() const {
-        return status_details;
-    }
+    /// Provides a reference to the internal PerfStats instance.
+    Core::PerfStats& GetPerfStats();
 
-    Loader::AppLoader& GetAppLoader() const {
-        return *app_loader;
-    }
+    /// Provides a constant reference to the internal PerfStats instance.
+    const Core::PerfStats& GetPerfStats() const;
+
+    /// Provides a reference to the frame limiter;
+    Core::FrameLimiter& FrameLimiter();
+
+    /// Provides a constant referent to the frame limiter
+    const Core::FrameLimiter& FrameLimiter() const;
+
+    /// Gets the name of the current game
+    Loader::ResultStatus GetGameName(std::string& out) const;
+
+    void SetStatus(ResultStatus new_status, const char* details);
+
+    const std::string& GetStatusDetails() const;
+
+    Loader::AppLoader& GetAppLoader() const;
 
     Service::SM::ServiceManager& ServiceManager();
     const Service::SM::ServiceManager& ServiceManager() const;
 
-    void SetGPUDebugContext(std::shared_ptr<Tegra::DebugContext> context) {
-        debug_context = std::move(context);
-    }
+    void SetGPUDebugContext(std::shared_ptr<Tegra::DebugContext> context);
 
-    std::shared_ptr<Tegra::DebugContext> GetGPUDebugContext() const {
-        return debug_context;
-    }
+    Tegra::DebugContext* GetGPUDebugContext() const;
+
+    void SetFilesystem(std::shared_ptr<FileSys::VfsFilesystem> vfs);
+
+    std::shared_ptr<FileSys::VfsFilesystem> GetFilesystem() const;
 
 private:
     System();
@@ -175,35 +223,16 @@ private:
 
     /**
      * Initialize the emulated system.
-     * @param emu_window Pointer to the host-system window used for video output and keyboard input.
-     * @param system_mode The system mode.
+     * @param emu_window Reference to the host-system window used for video output and keyboard
+     *                   input.
      * @return ResultStatus code, indicating if the operation succeeded.
      */
-    ResultStatus Init(EmuWindow* emu_window, u32 system_mode);
+    ResultStatus Init(Frontend::EmuWindow& emu_window);
 
-    /// AppLoader used to load the current executing application
-    std::unique_ptr<Loader::AppLoader> app_loader;
-    std::unique_ptr<Tegra::GPU> gpu_core;
-    std::shared_ptr<Tegra::DebugContext> debug_context;
-    Kernel::SharedPtr<Kernel::Process> current_process;
-    std::shared_ptr<CpuBarrier> cpu_barrier;
-    std::array<std::shared_ptr<Cpu>, NUM_CPU_CORES> cpu_cores;
-    std::array<std::unique_ptr<std::thread>, NUM_CPU_CORES - 1> cpu_core_threads;
-    size_t active_core{}; ///< Active core, only used in single thread mode
-
-    /// Service manager
-    std::shared_ptr<Service::SM::ServiceManager> service_manager;
-
-    /// Telemetry session for this emulation session
-    std::unique_ptr<Core::TelemetrySession> telemetry_session;
+    struct Impl;
+    std::unique_ptr<Impl> impl;
 
     static System s_instance;
-
-    ResultStatus status = ResultStatus::Success;
-    std::string status_details = "";
-
-    /// Map of guest threads to CPU cores
-    std::map<std::thread::id, std::shared_ptr<Cpu>> thread_to_cpu;
 };
 
 inline ARM_Interface& CurrentArmInterface() {

src/core/core_cpu.cpp

@@ -12,9 +12,9 @@
 #include "core/arm/unicorn/arm_unicorn.h"
 #include "core/core_cpu.h"
 #include "core/core_timing.h"
-#include "core/hle/kernel/kernel.h"
 #include "core/hle/kernel/scheduler.h"
 #include "core/hle/kernel/thread.h"
+#include "core/hle/lock.h"
 #include "core/settings.h"
 
 namespace Core {
@@ -48,14 +48,15 @@ bool CpuBarrier::Rendezvous() {
     return false;
 }
 
-Cpu::Cpu(std::shared_ptr<CpuBarrier> cpu_barrier, size_t core_index)
+Cpu::Cpu(std::shared_ptr<ExclusiveMonitor> exclusive_monitor,
+         std::shared_ptr<CpuBarrier> cpu_barrier, size_t core_index)
     : cpu_barrier{std::move(cpu_barrier)}, core_index{core_index} {
 
     if (Settings::values.use_cpu_jit) {
 #ifdef ARCHITECTURE_x86_64
-        arm_interface = std::make_shared<ARM_Dynarmic>();
+        arm_interface = std::make_shared<ARM_Dynarmic>(exclusive_monitor, core_index);
 #else
-        cpu_core = std::make_shared<ARM_Unicorn>();
+        arm_interface = std::make_shared<ARM_Unicorn>();
         LOG_WARNING(Core, "CPU JIT requested, but Dynarmic not available");
 #endif
     } else {
@@ -65,6 +66,18 @@ Cpu::Cpu(std::shared_ptr<CpuBarrier> cpu_barrier, size_t core_index)
     scheduler = std::make_shared<Kernel::Scheduler>(arm_interface.get());
 }
 
+std::shared_ptr<ExclusiveMonitor> Cpu::MakeExclusiveMonitor(size_t num_cores) {
+    if (Settings::values.use_cpu_jit) {
+#ifdef ARCHITECTURE_x86_64
+        return std::make_shared<DynarmicExclusiveMonitor>(num_cores);
+#else
+        return nullptr; // TODO(merry): Passthrough exclusive monitor
+#endif
+    } else {
+        return nullptr; // TODO(merry): Passthrough exclusive monitor
+    }
+}
+
 void Cpu::RunLoop(bool tight_loop) {
     // Wait for all other CPU cores to complete the previous slice, such that they run in lock-step
     if (!cpu_barrier->Rendezvous()) {
@@ -78,6 +91,7 @@ void Cpu::RunLoop(bool tight_loop) {
         LOG_TRACE(Core, "Core-{} idling", core_index);
 
         if (IsMainCore()) {
+            // TODO(Subv): Only let CoreTiming idle if all 4 cores are idling.
             CoreTiming::Idle();
             CoreTiming::Advance();
         }
@@ -113,6 +127,8 @@ void Cpu::Reschedule() {
     }
 
     reschedule_pending = false;
+    // Lock the global kernel mutex when we manipulate the HLE state
+    std::lock_guard<std::recursive_mutex> lock(HLE::g_hle_lock);
     scheduler->Reschedule();
 }
 

src/core/core_cpu.h

@@ -10,8 +10,7 @@
 #include <mutex>
 #include <string>
 #include "common/common_types.h"
-
-class ARM_Interface;
+#include "core/arm/exclusive_monitor.h"
 
 namespace Kernel {
 class Scheduler;
@@ -19,6 +18,8 @@ class Scheduler;
 
 namespace Core {
 
+class ARM_Interface;
+
 constexpr unsigned NUM_CPU_CORES{4};
 
 class CpuBarrier {
@@ -40,7 +41,8 @@ private:
 
 class Cpu {
 public:
-    Cpu(std::shared_ptr<CpuBarrier> cpu_barrier, size_t core_index);
+    Cpu(std::shared_ptr<ExclusiveMonitor> exclusive_monitor,
+        std::shared_ptr<CpuBarrier> cpu_barrier, size_t core_index);
 
     void RunLoop(bool tight_loop = true);
 
@@ -64,6 +66,12 @@ public:
         return core_index == 0;
     }
 
+    size_t CoreIndex() const {
+        return core_index;
+    }
+
+    static std::shared_ptr<ExclusiveMonitor> MakeExclusiveMonitor(size_t num_cores);
+
 private:
     void Reschedule();
 
@@ -71,7 +79,7 @@ private:
     std::shared_ptr<CpuBarrier> cpu_barrier;
     std::shared_ptr<Kernel::Scheduler> scheduler;
 
-    bool reschedule_pending{};
+    std::atomic<bool> reschedule_pending = false;
     size_t core_index;
 };
 

src/core/core_timing.cpp

@@ -5,17 +5,15 @@
 #include "core/core_timing.h"
 
 #include <algorithm>
-#include <cinttypes>
-#include <limits>
 #include <mutex>
 #include <string>
 #include <tuple>
 #include <unordered_map>
 #include <vector>
 #include "common/assert.h"
-#include "common/logging/log.h"
 #include "common/thread.h"
 #include "common/threadsafe_queue.h"
+#include "core/core_timing_util.h"
 
 namespace CoreTiming {
 
@@ -58,8 +56,10 @@ static u64 event_fifo_id;
 // to the event_queue by the emu thread
 static Common::MPSCQueue<Event, false> ts_queue;
 
+// the queue for unscheduling the events from other threads threadsafe
+static Common::MPSCQueue<std::pair<const EventType*, u64>, false> unschedule_queue;
+
 constexpr int MAX_SLICE_LENGTH = 20000;
-constexpr u64 MAX_VALUE_TO_MULTIPLY = std::numeric_limits<s64>::max() / BASE_CLOCK_RATE;
 
 static s64 idled_cycles;
 
@@ -72,54 +72,6 @@ static EventType* ev_lost = nullptr;
 
 static void EmptyTimedCallback(u64 userdata, s64 cyclesLate) {}
 
-s64 usToCycles(s64 us) {
-    if (us / 1000000 > MAX_VALUE_TO_MULTIPLY) {
-        LOG_ERROR(Core_Timing, "Integer overflow, use max value");
-        return std::numeric_limits<s64>::max();
-    }
-    if (us > MAX_VALUE_TO_MULTIPLY) {
-        LOG_DEBUG(Core_Timing, "Time very big, do rounding");
-        return BASE_CLOCK_RATE * (us / 1000000);
-    }
-    return (BASE_CLOCK_RATE * us) / 1000000;
-}
-
-s64 usToCycles(u64 us) {
-    if (us / 1000000 > MAX_VALUE_TO_MULTIPLY) {
-        LOG_ERROR(Core_Timing, "Integer overflow, use max value");
-        return std::numeric_limits<s64>::max();
-    }
-    if (us > MAX_VALUE_TO_MULTIPLY) {
-        LOG_DEBUG(Core_Timing, "Time very big, do rounding");
-        return BASE_CLOCK_RATE * static_cast<s64>(us / 1000000);
-    }
-    return (BASE_CLOCK_RATE * static_cast<s64>(us)) / 1000000;
-}
-
-s64 nsToCycles(s64 ns) {
-    if (ns / 1000000000 > MAX_VALUE_TO_MULTIPLY) {
-        LOG_ERROR(Core_Timing, "Integer overflow, use max value");
-        return std::numeric_limits<s64>::max();
-    }
-    if (ns > MAX_VALUE_TO_MULTIPLY) {
-        LOG_DEBUG(Core_Timing, "Time very big, do rounding");
-        return BASE_CLOCK_RATE * (ns / 1000000000);
-    }
-    return (BASE_CLOCK_RATE * ns) / 1000000000;
-}
-
-s64 nsToCycles(u64 ns) {
-    if (ns / 1000000000 > MAX_VALUE_TO_MULTIPLY) {
-        LOG_ERROR(Core_Timing, "Integer overflow, use max value");
-        return std::numeric_limits<s64>::max();
-    }
-    if (ns > MAX_VALUE_TO_MULTIPLY) {
-        LOG_DEBUG(Core_Timing, "Time very big, do rounding");
-        return BASE_CLOCK_RATE * (static_cast<s64>(ns) / 1000000000);
-    }
-    return (BASE_CLOCK_RATE * static_cast<s64>(ns)) / 1000000000;
-}
-
 EventType* RegisterEvent(const std::string& name, TimedCallback callback) {
     // check for existing type with same name.
     // we want event type names to remain unique so that we can use them for serialization.
@@ -186,13 +138,11 @@ void ClearPendingEvents() {
 void ScheduleEvent(s64 cycles_into_future, const EventType* event_type, u64 userdata) {
     ASSERT(event_type != nullptr);
     s64 timeout = GetTicks() + cycles_into_future;
-
     // If this event needs to be scheduled before the next advance(), force one early
     if (!is_global_timer_sane)
         ForceExceptionCheck(cycles_into_future);
-
     event_queue.emplace_back(Event{timeout, event_fifo_id++, userdata, event_type});
-    std::push_heap(event_queue.begin(), event_queue.end(), std::greater<Event>());
+    std::push_heap(event_queue.begin(), event_queue.end(), std::greater<>());
 }
 
 void ScheduleEventThreadsafe(s64 cycles_into_future, const EventType* event_type, u64 userdata) {
@@ -207,10 +157,14 @@ void UnscheduleEvent(const EventType* event_type, u64 userdata) {
     // Removing random items breaks the invariant so we have to re-establish it.
     if (itr != event_queue.end()) {
         event_queue.erase(itr, event_queue.end());
-        std::make_heap(event_queue.begin(), event_queue.end(), std::greater<Event>());
+        std::make_heap(event_queue.begin(), event_queue.end(), std::greater<>());
     }
 }
 
+void UnscheduleEventThreadsafe(const EventType* event_type, u64 userdata) {
+    unschedule_queue.Push(std::make_pair(event_type, userdata));
+}
+
 void RemoveEvent(const EventType* event_type) {
     auto itr = std::remove_if(event_queue.begin(), event_queue.end(),
                               [&](const Event& e) { return e.type == event_type; });
@@ -218,7 +172,7 @@ void RemoveEvent(const EventType* event_type) {
     // Removing random items breaks the invariant so we have to re-establish it.
     if (itr != event_queue.end()) {
         event_queue.erase(itr, event_queue.end());
-        std::make_heap(event_queue.begin(), event_queue.end(), std::greater<Event>());
+        std::make_heap(event_queue.begin(), event_queue.end(), std::greater<>());
     }
 }
 
@@ -241,12 +195,15 @@ void MoveEvents() {
     for (Event ev; ts_queue.Pop(ev);) {
         ev.fifo_order = event_fifo_id++;
         event_queue.emplace_back(std::move(ev));
-        std::push_heap(event_queue.begin(), event_queue.end(), std::greater<Event>());
+        std::push_heap(event_queue.begin(), event_queue.end(), std::greater<>());
     }
 }
 
 void Advance() {
     MoveEvents();
+    for (std::pair<const EventType*, u64> ev; unschedule_queue.Pop(ev);) {
+        UnscheduleEvent(ev.first, ev.second);
+    }
 
     int cycles_executed = slice_length - downcount;
     global_timer += cycles_executed;
@@ -256,7 +213,7 @@ void Advance() {
 
     while (!event_queue.empty() && event_queue.front().time <= global_timer) {
         Event evt = std::move(event_queue.front());
-        std::pop_heap(event_queue.begin(), event_queue.end(), std::greater<Event>());
+        std::pop_heap(event_queue.begin(), event_queue.end(), std::greater<>());
         event_queue.pop_back();
         evt.type->callback(evt.userdata, static_cast<int>(global_timer - evt.time));
     }
@@ -277,8 +234,8 @@ void Idle() {
     downcount = 0;
 }
 
-u64 GetGlobalTimeUs() {
-    return GetTicks() * 1000000 / BASE_CLOCK_RATE;
+std::chrono::microseconds GetGlobalTimeUs() {
+    return std::chrono::microseconds{GetTicks() * 1000000 / BASE_CLOCK_RATE};
 }
 
 int GetDowncount() {

src/core/core_timing.h

@@ -17,64 +17,16 @@
  *   ScheduleEvent(periodInCycles - cyclesLate, callback, "whatever")
  */
 
+#include <chrono>
 #include <functional>
 #include <string>
 #include "common/common_types.h"
 
 namespace CoreTiming {
 
-// The below clock rate is based on Switch's clockspeed being widely known as 1.020GHz
-// The exact value used is of course unverified.
-constexpr u64 BASE_CLOCK_RATE = 1019215872; // Switch clock speed is 1020MHz un/docked
+struct EventType;
 
-inline s64 msToCycles(int ms) {
-    // since ms is int there is no way to overflow
-    return BASE_CLOCK_RATE * static_cast<s64>(ms) / 1000;
-}
-
-inline s64 msToCycles(float ms) {
-    return static_cast<s64>(BASE_CLOCK_RATE * (0.001f) * ms);
-}
-
-inline s64 msToCycles(double ms) {
-    return static_cast<s64>(BASE_CLOCK_RATE * (0.001) * ms);
-}
-
-inline s64 usToCycles(float us) {
-    return static_cast<s64>(BASE_CLOCK_RATE * (0.000001f) * us);
-}
-
-inline s64 usToCycles(int us) {
-    return (BASE_CLOCK_RATE * static_cast<s64>(us) / 1000000);
-}
-
-s64 usToCycles(s64 us);
-
-s64 usToCycles(u64 us);
-
-inline s64 nsToCycles(float ns) {
-    return static_cast<s64>(BASE_CLOCK_RATE * (0.000000001f) * ns);
-}
-
-inline s64 nsToCycles(int ns) {
-    return BASE_CLOCK_RATE * static_cast<s64>(ns) / 1000000000;
-}
-
-s64 nsToCycles(s64 ns);
-
-s64 nsToCycles(u64 ns);
-
-inline u64 cyclesToNs(s64 cycles) {
-    return cycles * 1000000000 / BASE_CLOCK_RATE;
-}
-
-inline s64 cyclesToUs(s64 cycles) {
-    return cycles * 1000000 / BASE_CLOCK_RATE;
-}
-
-inline u64 cyclesToMs(s64 cycles) {
-    return cycles * 1000 / BASE_CLOCK_RATE;
-}
+using TimedCallback = std::function<void(u64 userdata, int cycles_late)>;
 
 /**
  * CoreTiming begins at the boundary of timing slice -1. An initial call to Advance() is
@@ -83,8 +35,6 @@ inline u64 cyclesToMs(s64 cycles) {
 void Init();
 void Shutdown();
 
-typedef std::function<void(u64 userdata, int cycles_late)> TimedCallback;
-
 /**
  * This should only be called from the emu thread, if you are calling it any other thread, you are
  * doing something evil
@@ -93,8 +43,6 @@ u64 GetTicks();
 u64 GetIdleTicks();
 void AddTicks(u64 ticks);
 
-struct EventType;
-
 /**
  * Returns the event_type identifier. if name is not unique, it will assert.
  */
@@ -117,6 +65,7 @@ void ScheduleEvent(s64 cycles_into_future, const EventType* event_type, u64 user
 void ScheduleEventThreadsafe(s64 cycles_into_future, const EventType* event_type, u64 userdata);
 
 void UnscheduleEvent(const EventType* event_type, u64 userdata);
+void UnscheduleEventThreadsafe(const EventType* event_type, u64 userdata);
 
 /// We only permit one event of each type in the queue at a time.
 void RemoveEvent(const EventType* event_type);
@@ -139,7 +88,7 @@ void ClearPendingEvents();
 
 void ForceExceptionCheck(s64 cycles);
 
-u64 GetGlobalTimeUs();
+std::chrono::microseconds GetGlobalTimeUs();
 
 int GetDowncount();
 

src/core/core_timing_util.cpp

@@ -0,0 +1,63 @@
+// Copyright 2008 Dolphin Emulator Project / 2017 Citra Emulator Project
+// Licensed under GPLv2+
+// Refer to the license.txt file included.
+
+#include "core/core_timing_util.h"
+
+#include <cinttypes>
+#include <limits>
+#include "common/logging/log.h"
+
+namespace CoreTiming {
+
+constexpr u64 MAX_VALUE_TO_MULTIPLY = std::numeric_limits<s64>::max() / BASE_CLOCK_RATE;
+
+s64 usToCycles(s64 us) {
+    if (us / 1000000 > MAX_VALUE_TO_MULTIPLY) {
+        LOG_ERROR(Core_Timing, "Integer overflow, use max value");
+        return std::numeric_limits<s64>::max();
+    }
+    if (us > MAX_VALUE_TO_MULTIPLY) {
+        LOG_DEBUG(Core_Timing, "Time very big, do rounding");
+        return BASE_CLOCK_RATE * (us / 1000000);
+    }
+    return (BASE_CLOCK_RATE * us) / 1000000;
+}
+
+s64 usToCycles(u64 us) {
+    if (us / 1000000 > MAX_VALUE_TO_MULTIPLY) {
+        LOG_ERROR(Core_Timing, "Integer overflow, use max value");
+        return std::numeric_limits<s64>::max();
+    }
+    if (us > MAX_VALUE_TO_MULTIPLY) {
+        LOG_DEBUG(Core_Timing, "Time very big, do rounding");
+        return BASE_CLOCK_RATE * static_cast<s64>(us / 1000000);
+    }
+    return (BASE_CLOCK_RATE * static_cast<s64>(us)) / 1000000;
+}
+
+s64 nsToCycles(s64 ns) {
+    if (ns / 1000000000 > MAX_VALUE_TO_MULTIPLY) {
+        LOG_ERROR(Core_Timing, "Integer overflow, use max value");
+        return std::numeric_limits<s64>::max();
+    }
+    if (ns > MAX_VALUE_TO_MULTIPLY) {
+        LOG_DEBUG(Core_Timing, "Time very big, do rounding");
+        return BASE_CLOCK_RATE * (ns / 1000000000);
+    }
+    return (BASE_CLOCK_RATE * ns) / 1000000000;
+}
+
+s64 nsToCycles(u64 ns) {
+    if (ns / 1000000000 > MAX_VALUE_TO_MULTIPLY) {
+        LOG_ERROR(Core_Timing, "Integer overflow, use max value");
+        return std::numeric_limits<s64>::max();
+    }
+    if (ns > MAX_VALUE_TO_MULTIPLY) {
+        LOG_DEBUG(Core_Timing, "Time very big, do rounding");
+        return BASE_CLOCK_RATE * (static_cast<s64>(ns) / 1000000000);
+    }
+    return (BASE_CLOCK_RATE * static_cast<s64>(ns)) / 1000000000;
+}
+
+} // namespace CoreTiming

src/core/core_timing_util.h

@@ -0,0 +1,64 @@
+// Copyright 2008 Dolphin Emulator Project / 2017 Citra Emulator Project
+// Licensed under GPLv2+
+// Refer to the license.txt file included.
+
+#pragma once
+
+#include "common/common_types.h"
+
+namespace CoreTiming {
+
+// The below clock rate is based on Switch's clockspeed being widely known as 1.020GHz
+// The exact value used is of course unverified.
+constexpr u64 BASE_CLOCK_RATE = 1019215872; // Switch clock speed is 1020MHz un/docked
+
+inline s64 msToCycles(int ms) {
+    // since ms is int there is no way to overflow
+    return BASE_CLOCK_RATE * static_cast<s64>(ms) / 1000;
+}
+
+inline s64 msToCycles(float ms) {
+    return static_cast<s64>(BASE_CLOCK_RATE * (0.001f) * ms);
+}
+
+inline s64 msToCycles(double ms) {
+    return static_cast<s64>(BASE_CLOCK_RATE * (0.001) * ms);
+}
+
+inline s64 usToCycles(float us) {
+    return static_cast<s64>(BASE_CLOCK_RATE * (0.000001f) * us);
+}
+
+inline s64 usToCycles(int us) {
+    return (BASE_CLOCK_RATE * static_cast<s64>(us) / 1000000);
+}
+
+s64 usToCycles(s64 us);
+
+s64 usToCycles(u64 us);
+
+inline s64 nsToCycles(float ns) {
+    return static_cast<s64>(BASE_CLOCK_RATE * (0.000000001f) * ns);
+}
+
+inline s64 nsToCycles(int ns) {
+    return BASE_CLOCK_RATE * static_cast<s64>(ns) / 1000000000;
+}
+
+s64 nsToCycles(s64 ns);
+
+s64 nsToCycles(u64 ns);
+
+inline u64 cyclesToNs(s64 cycles) {
+    return cycles * 1000000000 / BASE_CLOCK_RATE;
+}
+
+inline s64 cyclesToUs(s64 cycles) {
+    return cycles * 1000000 / BASE_CLOCK_RATE;
+}
+
+inline u64 cyclesToMs(s64 cycles) {
+    return cycles * 1000 / BASE_CLOCK_RATE;
+}
+
+} // namespace CoreTiming

src/core/crypto/aes_util.cpp

@@ -0,0 +1,112 @@
+// Copyright 2018 yuzu emulator team
+// Licensed under GPLv2 or any later version
+// Refer to the license.txt file included.
+
+#include <mbedtls/cipher.h>
+#include "common/assert.h"
+#include "common/logging/log.h"
+#include "core/crypto/aes_util.h"
+#include "core/crypto/key_manager.h"
+
+namespace Core::Crypto {
+namespace {
+std::vector<u8> CalculateNintendoTweak(size_t sector_id) {
+    std::vector<u8> out(0x10);
+    for (size_t i = 0xF; i <= 0xF; --i) {
+        out[i] = sector_id & 0xFF;
+        sector_id >>= 8;
+    }
+    return out;
+}
+} // Anonymous namespace
+
+static_assert(static_cast<size_t>(Mode::CTR) == static_cast<size_t>(MBEDTLS_CIPHER_AES_128_CTR),
+              "CTR has incorrect value.");
+static_assert(static_cast<size_t>(Mode::ECB) == static_cast<size_t>(MBEDTLS_CIPHER_AES_128_ECB),
+              "ECB has incorrect value.");
+static_assert(static_cast<size_t>(Mode::XTS) == static_cast<size_t>(MBEDTLS_CIPHER_AES_128_XTS),
+              "XTS has incorrect value.");
+
+// Structure to hide mbedtls types from header file
+struct CipherContext {
+    mbedtls_cipher_context_t encryption_context;
+    mbedtls_cipher_context_t decryption_context;
+};
+
+template <typename Key, size_t KeySize>
+Crypto::AESCipher<Key, KeySize>::AESCipher(Key key, Mode mode)
+    : ctx(std::make_unique<CipherContext>()) {
+    mbedtls_cipher_init(&ctx->encryption_context);
+    mbedtls_cipher_init(&ctx->decryption_context);
+
+    ASSERT_MSG((mbedtls_cipher_setup(
+                    &ctx->encryption_context,
+                    mbedtls_cipher_info_from_type(static_cast<mbedtls_cipher_type_t>(mode))) ||
+                mbedtls_cipher_setup(
+                    &ctx->decryption_context,
+                    mbedtls_cipher_info_from_type(static_cast<mbedtls_cipher_type_t>(mode)))) == 0,
+               "Failed to initialize mbedtls ciphers.");
+
+    ASSERT(
+        !mbedtls_cipher_setkey(&ctx->encryption_context, key.data(), KeySize * 8, MBEDTLS_ENCRYPT));
+    ASSERT(
+        !mbedtls_cipher_setkey(&ctx->decryption_context, key.data(), KeySize * 8, MBEDTLS_DECRYPT));
+    //"Failed to set key on mbedtls ciphers.");
+}
+
+template <typename Key, size_t KeySize>
+AESCipher<Key, KeySize>::~AESCipher() {
+    mbedtls_cipher_free(&ctx->encryption_context);
+    mbedtls_cipher_free(&ctx->decryption_context);
+}
+
+template <typename Key, size_t KeySize>
+void AESCipher<Key, KeySize>::SetIV(std::vector<u8> iv) {
+    ASSERT_MSG((mbedtls_cipher_set_iv(&ctx->encryption_context, iv.data(), iv.size()) ||
+                mbedtls_cipher_set_iv(&ctx->decryption_context, iv.data(), iv.size())) == 0,
+               "Failed to set IV on mbedtls ciphers.");
+}
+
+template <typename Key, size_t KeySize>
+void AESCipher<Key, KeySize>::Transcode(const u8* src, size_t size, u8* dest, Op op) const {
+    auto* const context = op == Op::Encrypt ? &ctx->encryption_context : &ctx->decryption_context;
+
+    mbedtls_cipher_reset(context);
+
+    size_t written = 0;
+    if (mbedtls_cipher_get_cipher_mode(context) == MBEDTLS_MODE_XTS) {
+        mbedtls_cipher_update(context, src, size, dest, &written);
+        if (written != size) {
+            LOG_WARNING(Crypto, "Not all data was decrypted requested={:016X}, actual={:016X}.",
+                        size, written);
+        }
+    } else {
+        const auto block_size = mbedtls_cipher_get_block_size(context);
+
+        for (size_t offset = 0; offset < size; offset += block_size) {
+            auto length = std::min<size_t>(block_size, size - offset);
+            mbedtls_cipher_update(context, src + offset, length, dest + offset, &written);
+            if (written != length) {
+                LOG_WARNING(Crypto, "Not all data was decrypted requested={:016X}, actual={:016X}.",
+                            length, written);
+            }
+        }
+    }
+
+    mbedtls_cipher_finish(context, nullptr, nullptr);
+}
+
+template <typename Key, size_t KeySize>
+void AESCipher<Key, KeySize>::XTSTranscode(const u8* src, size_t size, u8* dest, size_t sector_id,
+                                           size_t sector_size, Op op) {
+    ASSERT_MSG(size % sector_size == 0, "XTS decryption size must be a multiple of sector size.");
+
+    for (size_t i = 0; i < size; i += sector_size) {
+        SetIV(CalculateNintendoTweak(sector_id++));
+        Transcode<u8, u8>(src + i, sector_size, dest + i, op);
+    }
+}
+
+template class AESCipher<Key128>;
+template class AESCipher<Key256>;
+} // namespace Core::Crypto

src/core/crypto/aes_util.h

@@ -0,0 +1,64 @@
+// Copyright 2018 yuzu emulator team
+// Licensed under GPLv2 or any later version
+// Refer to the license.txt file included.
+
+#pragma once
+
+#include <memory>
+#include <type_traits>
+#include <vector>
+#include "common/common_types.h"
+#include "core/file_sys/vfs.h"
+
+namespace Core::Crypto {
+
+struct CipherContext;
+
+enum class Mode {
+    CTR = 11,
+    ECB = 2,
+    XTS = 70,
+};
+
+enum class Op {
+    Encrypt,
+    Decrypt,
+};
+
+template <typename Key, size_t KeySize = sizeof(Key)>
+class AESCipher {
+    static_assert(std::is_same_v<Key, std::array<u8, KeySize>>, "Key must be std::array of u8.");
+    static_assert(KeySize == 0x10 || KeySize == 0x20, "KeySize must be 128 or 256.");
+
+public:
+    AESCipher(Key key, Mode mode);
+
+    ~AESCipher();
+
+    void SetIV(std::vector<u8> iv);
+
+    template <typename Source, typename Dest>
+    void Transcode(const Source* src, size_t size, Dest* dest, Op op) const {
+        static_assert(std::is_trivially_copyable_v<Source> && std::is_trivially_copyable_v<Dest>,
+                      "Transcode source and destination types must be trivially copyable.");
+        Transcode(reinterpret_cast<const u8*>(src), size, reinterpret_cast<u8*>(dest), op);
+    }
+
+    void Transcode(const u8* src, size_t size, u8* dest, Op op) const;
+
+    template <typename Source, typename Dest>
+    void XTSTranscode(const Source* src, size_t size, Dest* dest, size_t sector_id,
+                      size_t sector_size, Op op) {
+        static_assert(std::is_trivially_copyable_v<Source> && std::is_trivially_copyable_v<Dest>,
+                      "XTSTranscode source and destination types must be trivially copyable.");
+        XTSTranscode(reinterpret_cast<const u8*>(src), size, reinterpret_cast<u8*>(dest), sector_id,
+                     sector_size, op);
+    }
+
+    void XTSTranscode(const u8* src, size_t size, u8* dest, size_t sector_id, size_t sector_size,
+                      Op op);
+
+private:
+    std::unique_ptr<CipherContext> ctx;
+};
+} // namespace Core::Crypto

src/core/crypto/ctr_encryption_layer.cpp

@@ -0,0 +1,54 @@
+// Copyright 2018 yuzu emulator team
+// Licensed under GPLv2 or any later version
+// Refer to the license.txt file included.
+
+#include <cstring>
+#include "common/assert.h"
+#include "core/crypto/ctr_encryption_layer.h"
+
+namespace Core::Crypto {
+
+CTREncryptionLayer::CTREncryptionLayer(FileSys::VirtualFile base_, Key128 key_, size_t base_offset)
+    : EncryptionLayer(std::move(base_)), base_offset(base_offset), cipher(key_, Mode::CTR),
+      iv(16, 0) {}
+
+size_t CTREncryptionLayer::Read(u8* data, size_t length, size_t offset) const {
+    if (length == 0)
+        return 0;
+
+    const auto sector_offset = offset & 0xF;
+    if (sector_offset == 0) {
+        UpdateIV(base_offset + offset);
+        std::vector<u8> raw = base->ReadBytes(length, offset);
+        cipher.Transcode(raw.data(), raw.size(), data, Op::Decrypt);
+        return raw.size();
+    }
+
+    // offset does not fall on block boundary (0x10)
+    std::vector<u8> block = base->ReadBytes(0x10, offset - sector_offset);
+    UpdateIV(base_offset + offset - sector_offset);
+    cipher.Transcode(block.data(), block.size(), block.data(), Op::Decrypt);
+    size_t read = 0x10 - sector_offset;
+
+    if (length + sector_offset < 0x10) {
+        std::memcpy(data, block.data() + sector_offset, std::min<u64>(length, read));
+        return std::min<u64>(length, read);
+    }
+    std::memcpy(data, block.data() + sector_offset, read);
+    return read + Read(data + read, length - read, offset + read);
+}
+
+void CTREncryptionLayer::SetIV(const std::vector<u8>& iv_) {
+    const auto length = std::min(iv_.size(), iv.size());
+    iv.assign(iv_.cbegin(), iv_.cbegin() + length);
+}
+
+void CTREncryptionLayer::UpdateIV(size_t offset) const {
+    offset >>= 4;
+    for (size_t i = 0; i < 8; ++i) {
+        iv[16 - i - 1] = offset & 0xFF;
+        offset >>= 8;
+    }
+    cipher.SetIV(iv);
+}
+} // namespace Core::Crypto