Fix buffer overlap checking skipping a page for stream score right expand

video_core: tune garbage collection aggressiveness

.ci/scripts/format/script.sh

@@ -10,7 +10,7 @@ if grep -nrI '\s$' src *.yml *.txt *.md Doxyfile .gitignore .gitmodules .ci* dis
 fi
 
 # Default clang-format points to default 3.5 version one
-CLANG_FORMAT=${CLANG_FORMAT:-clang-format-12}
+CLANG_FORMAT=${CLANG_FORMAT:-clang-format-15}
 $CLANG_FORMAT --version
 
 if [ "$TRAVIS_EVENT_TYPE" = "pull_request" ]; then

.ci/scripts/linux/docker.sh

@@ -22,6 +22,7 @@ cmake .. \
       -DUSE_DISCORD_PRESENCE=ON \
       -DYUZU_ENABLE_COMPATIBILITY_REPORTING=${ENABLE_COMPATIBILITY_REPORTING:-"OFF"} \
       -DYUZU_USE_BUNDLED_FFMPEG=ON \
+      -DYUZU_ENABLE_LTO=ON \
       -GNinja
 
 ninja

.ci/scripts/merge/apply-patches-by-label.py

@@ -2,15 +2,12 @@
 # SPDX-License-Identifier: GPL-2.0-or-later
 
 # Download all pull requests as patches that match a specific label
-# Usage: python download-patches-by-label.py <Label to Match> <Root Path Folder to DL to>
+# Usage: python apply-patches-by-label.py <Label to Match>
 
-import requests, sys, json, urllib3.request, shutil, subprocess, os, traceback
+import json, requests, subprocess, sys, traceback
 
 tagline = sys.argv[2]
 
-http = urllib3.PoolManager()
-dl_list = {}
-
 def check_individual(labels):
     for label in labels:
         if (label["name"] == sys.argv[1]):
@@ -18,8 +15,9 @@ def check_individual(labels):
     return False
 
 def do_page(page):
-    url = 'https://api.github.com/repos/yuzu-emu/yuzu/pulls?page=%s' % page
+    url = f"https://api.github.com/repos/yuzu-emu/yuzu/pulls?page={page}"
     response = requests.get(url)
+    response.raise_for_status()
     if (response.ok):
         j = json.loads(response.content)
         if j == []:
@@ -27,13 +25,13 @@ def do_page(page):
         for pr in j:
             if (check_individual(pr["labels"])):
                 pn = pr["number"]
-                print("Matched PR# %s" % pn)
-                print(subprocess.check_output(["git", "fetch", "https://github.com/yuzu-emu/yuzu.git", "pull/%s/head:pr-%s" % (pn, pn), "-f", "--no-recurse-submodules"]))
-                print(subprocess.check_output(["git", "merge", "--squash", "pr-%s" % pn]))
-                print(subprocess.check_output(["git", "commit", "-m\"Merge %s PR %s\"" % (tagline, pn)]))
+                print(f"Matched PR# {pn}")
+                print(subprocess.check_output(["git", "fetch", "https://github.com/yuzu-emu/yuzu.git", f"pull/{pn}/head:pr-{pn}", "-f", "--no-recurse-submodules"]))
+                print(subprocess.check_output(["git", "merge", "--squash", f"pr-{pn}"]))
+                print(subprocess.check_output(["git", "commit", f"-m\"Merge {tagline} PR {pn}\""]))
 
 try:
-    for i in range(1,30):
+    for i in range(1,10):
         do_page(i)
 except:
     traceback.print_exc(file=sys.stdout)

.ci/scripts/windows/scan_dll.py

@@ -40,7 +40,7 @@ def parse_imports(file_name):
 
 def parse_imports_recursive(file_name, path_list=[]):
     q = queue.Queue()  # create a FIFO queue
-    # file_name can be a string or a list for the convience
+    # file_name can be a string or a list for the convenience
     if isinstance(file_name, str):
         q.put(file_name)
     elif isinstance(file_name, list):

.ci/templates/build-msvc.yml

@@ -9,7 +9,7 @@ parameters:
 steps:
 - script: choco install vulkan-sdk
   displayName: 'Install vulkan-sdk'
-- script: refreshenv && mkdir build && cd build && cmake -E env CXXFLAGS="/Gw /GA /Gr /Ob2" cmake -G "Visual Studio 17 2022" -A x64 -DCMAKE_INTERPROCEDURAL_OPTIMIZATION=ON -DCMAKE_POLICY_DEFAULT_CMP0069=NEW -DYUZU_USE_BUNDLED_QT=1 -DYUZU_USE_BUNDLED_SDL2=1 -DYUZU_USE_QT_WEB_ENGINE=ON -DENABLE_COMPATIBILITY_LIST_DOWNLOAD=ON -DYUZU_ENABLE_COMPATIBILITY_REPORTING=${COMPAT} -DYUZU_TESTS=OFF -DUSE_DISCORD_PRESENCE=ON -DENABLE_QT_TRANSLATION=ON -DDISPLAY_VERSION=${{ parameters['version'] }} -DCMAKE_BUILD_TYPE=Release -DYUZU_CRASH_DUMPS=ON .. && cd ..
+- script: refreshenv && mkdir build && cd build && cmake -E env CXXFLAGS="/Gw /GA /Gr /Ob2" cmake -G "Visual Studio 17 2022" -A x64 -DCMAKE_POLICY_DEFAULT_CMP0069=NEW -DYUZU_ENABLE_LTO=ON -DYUZU_USE_BUNDLED_QT=1 -DYUZU_USE_BUNDLED_SDL2=1 -DYUZU_USE_QT_WEB_ENGINE=ON -DENABLE_COMPATIBILITY_LIST_DOWNLOAD=ON -DYUZU_ENABLE_COMPATIBILITY_REPORTING=${COMPAT} -DYUZU_TESTS=OFF -DUSE_DISCORD_PRESENCE=ON -DENABLE_QT_TRANSLATION=ON -DDISPLAY_VERSION=${{ parameters['version'] }} -DCMAKE_BUILD_TYPE=Release -DYUZU_CRASH_DUMPS=ON .. && cd ..
   displayName: 'Configure CMake'
 - task: MSBuild@1
   displayName: 'Build'

.codespellrc

@@ -0,0 +1,6 @@
+; SPDX-FileCopyrightText: 2023 yuzu Emulator Project
+; SPDX-License-Identifier: GPL-2.0-or-later
+
+[codespell]
+skip = ./.git,./build,./dist,./Doxyfile,./externals,./LICENSES
+ignore-words-list = aci,allright,ba,deques,froms,hda,inout,lod,masia,nam,nax,nd,pullrequests,pullrequest,te,transfered,unstall,uscaled,zink

.github/workflows/codespell.yml

@@ -0,0 +1,17 @@
+# SPDX-FileCopyrightText: 2023 yuzu Emulator Project
+# SPDX-License-Identifier: GPL-2.0-or-later
+# GitHub Action to automate the identification of common misspellings in text files.
+# https://github.com/codespell-project/actions-codespell
+# https://github.com/codespell-project/codespell
+name: codespell
+on: pull_request
+permissions: {}
+jobs:
+  codespell:
+    name: Check for spelling errors
+    runs-on: ubuntu-latest
+    steps:
+      - uses: actions/checkout@v3
+        with:
+          persist-credentials: false
+      - uses: codespell-project/actions-codespell@master

.gitmodules

@@ -13,9 +13,6 @@
 [submodule "dynarmic"]
     path = externals/dynarmic
     url = https://github.com/MerryMage/dynarmic.git
-[submodule "libressl"]
-    path = externals/libressl
-    url = https://github.com/citra-emu/ext-libressl-portable.git
 [submodule "libusb"]
 	path = externals/libusb/libusb
 	url = https://github.com/libusb/libusb.git

.lgtm.yml

@@ -1,13 +0,0 @@
-# SPDX-FileCopyrightText: 2020 yuzu Emulator Project
-# SPDX-License-Identifier: GPL-2.0-or-later
-
-path_classifiers:
-  library: "externals"
-extraction:
-  cpp:
-    prepare:
-      packages:
-      - "libsdl2-dev"
-      - "qtmultimedia5-dev"
-      - "libtbb-dev"
-      - "libjack-jackd2-dev"

CMakeLists.txt

@@ -3,17 +3,14 @@
 
 cmake_minimum_required(VERSION 3.22)
 
-# Dynarmic has cmake_minimum_required(3.12) and we may want to override
-# some of its variables, which is only possible in 3.13+
-set(CMAKE_POLICY_DEFAULT_CMP0077 NEW)
+project(yuzu)
 
 list(APPEND CMAKE_MODULE_PATH "${CMAKE_CURRENT_SOURCE_DIR}/CMakeModules")
 list(APPEND CMAKE_MODULE_PATH "${CMAKE_CURRENT_SOURCE_DIR}/externals/cmake-modules")
-list(APPEND CMAKE_MODULE_PATH "${CMAKE_CURRENT_SOURCE_DIR}/externals/find-modules")
+
 include(DownloadExternals)
 include(CMakeDependentOption)
-
-project(yuzu)
+include(CTest)
 
 # Set bundled sdl2/qt as dependent options.
 # OFF by default, but if ENABLE_SDL2 and MSVC are true then ON
@@ -47,7 +44,7 @@ option(ENABLE_CUBEB "Enables the cubeb audio backend" ON)
 
 option(USE_DISCORD_PRESENCE "Enables Discord Rich Presence" OFF)
 
-option(YUZU_TESTS "Compile tests" ON)
+option(YUZU_TESTS "Compile tests" "${BUILD_TESTING}")
 
 option(YUZU_USE_PRECOMPILED_HEADERS "Use precompiled headers" ON)
 
@@ -59,6 +56,8 @@ option(YUZU_USE_BUNDLED_VCPKG "Use vcpkg for yuzu dependencies" "${MSVC}")
 
 option(YUZU_CHECK_SUBMODULES "Check if submodules are present" ON)
 
+option(YUZU_ENABLE_LTO "Enable link-time optimization" OFF)
+
 CMAKE_DEPENDENT_OPTION(YUZU_USE_FASTER_LD "Check if a faster linker is available" ON "NOT WIN32" OFF)
 
 if (YUZU_USE_BUNDLED_VCPKG)
@@ -68,6 +67,9 @@ if (YUZU_USE_BUNDLED_VCPKG)
     if (YUZU_CRASH_DUMPS)
         list(APPEND VCPKG_MANIFEST_FEATURES "dbghelp")
     endif()
+    if (ENABLE_WEB_SERVICE)
+        list(APPEND VCPKG_MANIFEST_FEATURES "web-service")
+    endif()
 
     include(${CMAKE_SOURCE_DIR}/externals/vcpkg/scripts/buildsystems/vcpkg.cmake)
 elseif(NOT "$ENV{VCPKG_TOOLCHAIN_FILE}" STREQUAL "")
@@ -207,59 +209,49 @@ set(CMAKE_RUNTIME_OUTPUT_DIRECTORY ${PROJECT_BINARY_DIR}/bin)
 # System imported libraries
 # =======================================================================
 
-find_package(enet 1.3)
+# Enforce the search mode of non-required packages for better and shorter failure messages
+find_package(Boost 1.79.0 REQUIRED context)
+find_package(enet 1.3 MODULE)
 find_package(fmt 9 REQUIRED)
-find_package(inih)
+find_package(inih 52 MODULE COMPONENTS INIReader)
+find_package(LLVM MODULE COMPONENTS Demangle)
 find_package(lz4 REQUIRED)
 find_package(nlohmann_json 3.8 REQUIRED)
-find_package(Opus 1.3)
+find_package(Opus 1.3 MODULE)
 find_package(ZLIB 1.2 REQUIRED)
 find_package(zstd 1.5 REQUIRED)
 
 if (NOT YUZU_USE_EXTERNAL_VULKAN_HEADERS)
-    find_package(Vulkan 1.3.238)
+    find_package(Vulkan 1.3.246 REQUIRED)
 endif()
 
 if (ENABLE_LIBUSB)
-    find_package(libusb 1.0.24)
+    find_package(libusb 1.0.24 MODULE)
 endif()
 
 if (ARCHITECTURE_x86 OR ARCHITECTURE_x86_64)
-    find_package(xbyak 6 QUIET)
+    find_package(xbyak 6 CONFIG)
 endif()
 
 if (ARCHITECTURE_x86_64 OR ARCHITECTURE_arm64)
-    find_package(dynarmic 6.4.0 QUIET)
+    find_package(dynarmic 6.4.0 CONFIG)
 endif()
 
 if (ENABLE_CUBEB)
-    find_package(cubeb QUIET)
+    find_package(cubeb CONFIG)
 endif()
 
 if (USE_DISCORD_PRESENCE)
-    find_package(DiscordRPC QUIET)
+    find_package(DiscordRPC MODULE)
 endif()
 
 if (ENABLE_WEB_SERVICE)
-    find_package(cpp-jwt 1.4 QUIET)
-    find_package(httplib 0.11 QUIET)
+    find_package(cpp-jwt 1.4 CONFIG)
+    find_package(httplib 0.12 MODULE COMPONENTS OpenSSL)
 endif()
 
 if (YUZU_TESTS)
-    find_package(Catch2 2.13.7 REQUIRED)
-endif()
-
-find_package(Boost 1.73.0 COMPONENTS context)
-if (Boost_FOUND)
-    set(Boost_LIBRARIES Boost::boost)
-    # Conditionally add Boost::context only if the found Boost package provides it
-    # The old version is missing Boost::context, so we want to avoid adding in that case
-    # The new version requires adding Boost::context to prevent linking issues
-    if (TARGET Boost::context)
-        list(APPEND Boost_LIBRARIES Boost::context)
-    endif()
-else()
-    message(FATAL_ERROR "Boost 1.73.0 or newer not found")
+    find_package(Catch2 3.0.1 REQUIRED)
 endif()
 
 # boost:asio has functions that require AcceptEx et al
@@ -352,12 +344,12 @@ if(ENABLE_QT)
                 find_package(PkgConfig REQUIRED)
                 pkg_check_modules(QT_DEP_GLU QUIET glu>=9.0.0)
                 if (NOT QT_DEP_GLU_FOUND)
-                    message(FATAL_ERROR "Qt bundled pacakge dependency `glu` not found. \
+                    message(FATAL_ERROR "Qt bundled package dependency `glu` not found. \
                         Perhaps `libglu1-mesa-dev` needs to be installed?")
                 endif()
                 pkg_check_modules(QT_DEP_MESA QUIET dri>=20.0.8)
                 if (NOT QT_DEP_MESA_FOUND)
-                    message(FATAL_ERROR "Qt bundled pacakge dependency `dri` not found. \
+                    message(FATAL_ERROR "Qt bundled package dependency `dri` not found. \
                         Perhaps `mesa-common-dev` needs to be installed?")
                 endif()
 
@@ -458,14 +450,6 @@ if (ENABLE_SDL2)
     endif()
 endif()
 
-# Reexport some targets that are named differently when using the upstream CmakeConfig
-# In order to ALIAS targets to a new name, they first need to be IMPORTED_GLOBAL
-# Dynarmic checks for target `boost` and so we want to make sure it can find it through our system instead of using their external
-if (TARGET Boost::boost)
-    set_target_properties(Boost::boost PROPERTIES IMPORTED_GLOBAL TRUE)
-    add_library(boost ALIAS Boost::boost)
-endif()
-
 # List of all FFmpeg components required
 set(FFmpeg_COMPONENTS
     avcodec
@@ -493,8 +477,8 @@ if (APPLE)
     find_library(COCOA_LIBRARY Cocoa)
     set(PLATFORM_LIBRARIES ${COCOA_LIBRARY} ${IOKIT_LIBRARY} ${COREVIDEO_LIBRARY})
 elseif (WIN32)
-    # WSAPoll and SHGetKnownFolderPath (AppData/Roaming) didn't exist before WinNT 6.x (Vista)
-    add_definitions(-D_WIN32_WINNT=0x0600 -DWINVER=0x0600)
+    # Target Windows 10
+    add_definitions(-D_WIN32_WINNT=0x0A00 -DWINVER=0x0A00)
     set(PLATFORM_LIBRARIES winmm ws2_32 iphlpapi)
     if (MINGW)
         # PSAPI is the Process Status API
@@ -515,7 +499,7 @@ endif()
 # against all the src files. This should be used before making a pull request.
 # =======================================================================
 
-set(CLANG_FORMAT_POSTFIX "-12")
+set(CLANG_FORMAT_POSTFIX "-15")
 find_program(CLANG_FORMAT
     NAMES clang-format${CLANG_FORMAT_POSTFIX}
           clang-format
@@ -581,11 +565,7 @@ function(create_target_directory_groups target_name)
 endfunction()
 
 # Prevent boost from linking against libs when building
-add_definitions(-DBOOST_ERROR_CODE_HEADER_ONLY
-    -DBOOST_SYSTEM_NO_LIB
-    -DBOOST_DATE_TIME_NO_LIB
-    -DBOOST_REGEX_NO_LIB
-)
+target_link_libraries(Boost::headers INTERFACE Boost::disable_autolinking)
 # Adjustments for MSVC + Ninja
 if (MSVC AND CMAKE_GENERATOR STREQUAL "Ninja")
     add_compile_options(
@@ -610,7 +590,6 @@ if (YUZU_USE_FASTER_LD AND CMAKE_CXX_COMPILER_ID STREQUAL "GNU")
     endif()
 endif()
 
-enable_testing()
 add_subdirectory(externals)
 add_subdirectory(src)
 

CMakeModules/FindFFmpeg.cmake

@@ -14,7 +14,7 @@
 #  FFmpeg_LIBRARIES: aggregate all the paths to the libraries
 #  FFmpeg_FOUND: True if all components have been found
 #
-# This module defines the following targets, which are prefered over variables:
+# This module defines the following targets, which are preferred over variables:
 #
 #  FFmpeg::<component>: Target to use <component> directly, with include path,
 #    library and dependencies set up. If you are using a static build, you are

CMakeModules/FindLLVM.cmake

@@ -0,0 +1,26 @@
+# SPDX-FileCopyrightText: 2023 Alexandre Bouvier <contact@amb.tf>
+#
+# SPDX-License-Identifier: GPL-3.0-or-later
+
+find_package(LLVM QUIET COMPONENTS CONFIG)
+if (LLVM_FOUND)
+    separate_arguments(LLVM_DEFINITIONS)
+    if (LLVMDemangle IN_LIST LLVM_AVAILABLE_LIBS)
+        set(LLVM_Demangle_FOUND TRUE)
+    endif()
+endif()
+
+include(FindPackageHandleStandardArgs)
+find_package_handle_standard_args(LLVM HANDLE_COMPONENTS CONFIG_MODE)
+
+if (LLVM_FOUND AND LLVM_Demangle_FOUND AND NOT TARGET LLVM::Demangle)
+    add_library(LLVM::Demangle INTERFACE IMPORTED)
+    target_compile_definitions(LLVM::Demangle INTERFACE ${LLVM_DEFINITIONS})
+    target_include_directories(LLVM::Demangle INTERFACE ${LLVM_INCLUDE_DIRS})
+    # prefer shared LLVM: https://github.com/llvm/llvm-project/issues/34593
+    # but use ugly hack because llvm_config doesn't support interface library
+    add_library(_dummy_lib SHARED EXCLUDE_FROM_ALL src/yuzu/main.cpp)
+    llvm_config(_dummy_lib USE_SHARED demangle)
+    get_target_property(LLVM_LIBRARIES _dummy_lib LINK_LIBRARIES)
+    target_link_libraries(LLVM::Demangle INTERFACE ${LLVM_LIBRARIES})
+endif()

CMakeModules/Findhttplib.cmake

@@ -5,14 +5,24 @@
 include(FindPackageHandleStandardArgs)
 
 find_package(httplib QUIET CONFIG)
-if (httplib_FOUND)
-    find_package_handle_standard_args(httplib CONFIG_MODE)
+if (httplib_CONSIDERED_CONFIGS)
+    find_package_handle_standard_args(httplib HANDLE_COMPONENTS CONFIG_MODE)
 else()
     find_package(PkgConfig QUIET)
     pkg_search_module(HTTPLIB QUIET IMPORTED_TARGET cpp-httplib)
+    if ("-DCPPHTTPLIB_OPENSSL_SUPPORT" IN_LIST HTTPLIB_CFLAGS_OTHER)
+        set(httplib_OpenSSL_FOUND TRUE)
+    endif()
+    if ("-DCPPHTTPLIB_ZLIB_SUPPORT" IN_LIST HTTPLIB_CFLAGS_OTHER)
+        set(httplib_ZLIB_FOUND TRUE)
+    endif()
+    if ("-DCPPHTTPLIB_BROTLI_SUPPORT" IN_LIST HTTPLIB_CFLAGS_OTHER)
+        set(httplib_Brotli_FOUND TRUE)
+    endif()
     find_package_handle_standard_args(httplib
         REQUIRED_VARS HTTPLIB_INCLUDEDIR
         VERSION_VAR HTTPLIB_VERSION
+        HANDLE_COMPONENTS
     )
 endif()
 

CMakeModules/Findinih.cmake

@@ -0,0 +1,27 @@
+# SPDX-FileCopyrightText: 2022 Alexandre Bouvier <contact@amb.tf>
+#
+# SPDX-License-Identifier: GPL-3.0-or-later
+
+find_package(PkgConfig QUIET)
+pkg_search_module(INIH QUIET IMPORTED_TARGET inih)
+if (INIReader IN_LIST inih_FIND_COMPONENTS)
+    pkg_search_module(INIREADER QUIET IMPORTED_TARGET INIReader)
+    if (INIREADER_FOUND)
+        set(inih_INIReader_FOUND TRUE)
+    endif()
+endif()
+
+include(FindPackageHandleStandardArgs)
+find_package_handle_standard_args(inih
+    REQUIRED_VARS INIH_LINK_LIBRARIES
+    VERSION_VAR INIH_VERSION
+    HANDLE_COMPONENTS
+)
+
+if (inih_FOUND AND NOT TARGET inih::inih)
+    add_library(inih::inih ALIAS PkgConfig::INIH)
+endif()
+
+if (inih_FOUND AND inih_INIReader_FOUND AND NOT TARGET inih::INIReader)
+    add_library(inih::INIReader ALIAS PkgConfig::INIREADER)
+endif()

CMakeModules/Findlz4.cmake

@@ -4,7 +4,7 @@
 include(FindPackageHandleStandardArgs)
 
 find_package(lz4 QUIET CONFIG)
-if (lz4_FOUND)
+if (lz4_CONSIDERED_CONFIGS)
     find_package_handle_standard_args(lz4 CONFIG_MODE)
 else()
     find_package(PkgConfig QUIET)

CMakeModules/Findzstd.cmake

@@ -4,7 +4,7 @@
 include(FindPackageHandleStandardArgs)
 
 find_package(zstd QUIET CONFIG)
-if (zstd_FOUND)
+if (zstd_CONSIDERED_CONFIGS)
     find_package_handle_standard_args(zstd CONFIG_MODE)
 else()
     find_package(PkgConfig QUIET)

LICENSES/LLVM-exception.txt

@@ -0,0 +1,15 @@
+---- LLVM Exceptions to the Apache 2.0 License ----
+
+   As an exception, if, as a result of your compiling your source code, portions
+   of this Software are embedded into an Object form of such source code, you
+   may redistribute such embedded portions in such Object form without complying
+   with the conditions of Sections 4(a), 4(b) and 4(d) of the License.
+
+   In addition, if you combine or link compiled forms of this Software with
+   software that is licensed under the GPLv2 ("Combined Software") and if a
+   court of competent jurisdiction determines that the patent provision (Section
+   3), the indemnity provision (Section 9) or other Section of the License
+   conflicts with the conditions of the GPLv2, you may retroactively and
+   prospectively choose to deem waived or otherwise exclude such Section(s) of
+   the License, but only in their entirety and only with respect to the Combined
+   Software.

README.md

@@ -83,5 +83,3 @@ If you wish to support us a different way, please join our [Discord](https://dis
 ## License
 
 yuzu is licensed under the GPLv3 (or any later version). Refer to the [LICENSE.txt](https://github.com/yuzu-emu/yuzu/blob/master/LICENSE.txt) file.
-
-The [Skyline-Emulator Team](https://github.com/skyline-emu/skyline) may choose to use the code from these contributors under the GPL-3.0-or-later OR MPL-2.0: [FernandoS27](https://github.com/FernandoS27), [lioncash](https://github.com/lioncash), [bunnei](https://github.com/bunnei), [ReinUsesLisp](https://github.com/ReinUsesLisp), [Morph1984](https://github.com/Morph1984), [ogniK5377](https://github.com/ogniK5377), [german77](https://github.com/german77), [ameerj](https://github.com/ameerj), [Kelebek1](https://github.com/Kelebek1) and [lat9nq](https://github.com/lat9nq)

dist/org.yuzu_emu.yuzu.metainfo.xml

@@ -53,10 +53,10 @@ SPDX-License-Identifier: CC0-1.0
   <developer_name>yuzu Emulator Team</developer_name>
   <content_rating type="oars-1.0"/>
   <screenshots>
-    <screenshot type="default">https://raw.githubusercontent.com/yuzu-emu/yuzu-emu.github.io/master/images/screenshots/001-Super%20Mario%20Odyssey%20.png</screenshot>
-    <screenshot>https://raw.githubusercontent.com/yuzu-emu/yuzu-emu.github.io/master/images/screenshots/004-The%20Legend%20of%20Zelda%20Skyward%20Sword%20HD.png</screenshot>
-    <screenshot>https://raw.githubusercontent.com/yuzu-emu/yuzu-emu.github.io/master/images/screenshots/007-Pokemon%20Sword.png</screenshot>
-    <screenshot>https://raw.githubusercontent.com/yuzu-emu/yuzu-emu.github.io/master/images/screenshots/010-Hyrule%20Warriors%20Age%20of%20Calamity.png</screenshot>
-    <screenshot>https://raw.githubusercontent.com/yuzu-emu/yuzu-emu.github.io/master/images/screenshots/039-Pok%C3%A9mon%20Mystery%20Dungeon%20Rescue%20Team%20DX.png.png.png</screenshot>
+    <screenshot type="default"><image>https://raw.githubusercontent.com/yuzu-emu/yuzu-emu.github.io/master/images/screenshots/001-Super%20Mario%20Odyssey%20.png</image></screenshot>
+    <screenshot><image>https://raw.githubusercontent.com/yuzu-emu/yuzu-emu.github.io/master/images/screenshots/004-The%20Legend%20of%20Zelda%20Skyward%20Sword%20HD.png</image></screenshot>
+    <screenshot><image>https://raw.githubusercontent.com/yuzu-emu/yuzu-emu.github.io/master/images/screenshots/007-Pokemon%20Sword.png</image></screenshot>
+    <screenshot><image>https://raw.githubusercontent.com/yuzu-emu/yuzu-emu.github.io/master/images/screenshots/010-Hyrule%20Warriors%20Age%20of%20Calamity.png</image></screenshot>
+    <screenshot><image>https://raw.githubusercontent.com/yuzu-emu/yuzu-emu.github.io/master/images/screenshots/039-Pok%C3%A9mon%20Mystery%20Dungeon%20Rescue%20Team%20DX.png.png.png</image></screenshot>
   </screenshots>
 </component>

dist/yuzu.manifest

@@ -36,12 +36,6 @@ SPDX-License-Identifier: GPL-2.0-or-later
     <application>
       <!-- Windows 10 -->
       <supportedOS Id="{8e0f7a12-bfb3-4fe8-b9a5-48fd50a15a9a}"/>
-      <!-- Windows 8.1 -->
-      <supportedOS Id="{1f676c76-80e1-4239-95bb-83d0f6d0da78}"/>
-      <!-- Windows 8 -->
-      <supportedOS Id="{4a2f28e3-53b9-4441-ba9c-d69d4a4a6e38}"/>
-      <!-- Windows 7 -->
-      <supportedOS Id="{35138b9a-5d96-4fbd-8e2d-a2440225f93a}"/>
     </application>
   </compatibility>
   <trustInfo

externals/CMakeLists.txt

@@ -1,9 +1,12 @@
 # SPDX-FileCopyrightText: 2016 Citra Emulator Project
 # SPDX-License-Identifier: GPL-2.0-or-later
 
-list(APPEND CMAKE_MODULE_PATH "${PROJECT_SOURCE_DIR}/CMakeModules")
-list(APPEND CMAKE_MODULE_PATH "${PROJECT_SOURCE_DIR}/externals/find-modules")
-include(DownloadExternals)
+# Dynarmic has cmake_minimum_required(3.12) and we may want to override
+# some of its variables, which is only possible in 3.13+
+set(CMAKE_POLICY_DEFAULT_CMP0077 NEW)
+
+# Disable tests in all externals supporting the standard option name
+set(BUILD_TESTING OFF)
 
 # xbyak
 if ((ARCHITECTURE_x86 OR ARCHITECTURE_x86_64) AND NOT TARGET xbyak::xbyak)
@@ -12,8 +15,7 @@ endif()
 
 # Dynarmic
 if ((ARCHITECTURE_x86_64 OR ARCHITECTURE_arm64) AND NOT TARGET dynarmic::dynarmic)
-    set(DYNARMIC_NO_BUNDLED_FMT ON)
-    set(DYNARMIC_IGNORE_ASSERTS ON CACHE BOOL "" FORCE)
+    set(DYNARMIC_IGNORE_ASSERTS ON)
     add_subdirectory(dynarmic EXCLUDE_FROM_ALL)
     add_library(dynarmic::dynarmic ALIAS dynarmic)
 endif()
@@ -60,10 +62,10 @@ if (YUZU_USE_EXTERNAL_SDL2)
             Locale Power Render)
         foreach(_SUB ${SDL_UNUSED_SUBSYSTEMS})
           string(TOUPPER ${_SUB} _OPT)
-          option(SDL_${_OPT} "" OFF)
+          set(SDL_${_OPT} OFF)
         endforeach()
 
-        option(HIDAPI "" ON)
+        set(HIDAPI ON)
     endif()
     set(SDL_STATIC ON)
     set(SDL_SHARED OFF)
@@ -83,7 +85,7 @@ endif()
 
 # Cubeb
 if (ENABLE_CUBEB AND NOT TARGET cubeb::cubeb)
-    set(BUILD_TESTS OFF CACHE BOOL "")
+    set(BUILD_TESTS OFF)
     add_subdirectory(cubeb EXCLUDE_FROM_ALL)
     add_library(cubeb::cubeb ALIAS cubeb)
 endif()
@@ -98,43 +100,18 @@ endif()
 # Sirit
 add_subdirectory(sirit EXCLUDE_FROM_ALL)
 
+# httplib
 if (ENABLE_WEB_SERVICE AND NOT TARGET httplib::httplib)
-    if (NOT WIN32)
-        find_package(OpenSSL 1.1)
-        if (OPENSSL_FOUND)
-            set(OPENSSL_LIBRARIES OpenSSL::SSL OpenSSL::Crypto)
-        endif()
-    endif()
-
-    if (WIN32 OR NOT OPENSSL_FOUND)
-        # LibreSSL
-        set(LIBRESSL_SKIP_INSTALL ON CACHE BOOL "")
-        set(OPENSSLDIR "/etc/ssl/")
-        add_subdirectory(libressl EXCLUDE_FROM_ALL)
-        target_include_directories(ssl INTERFACE ./libressl/include)
-        target_compile_definitions(ssl PRIVATE -DHAVE_INET_NTOP)
-        get_directory_property(OPENSSL_LIBRARIES
-            DIRECTORY libressl
-            DEFINITION OPENSSL_LIBS)
-    endif()
-
-    # httplib
-    add_library(httplib INTERFACE)
-    target_include_directories(httplib INTERFACE ./cpp-httplib)
-    target_compile_definitions(httplib INTERFACE -DCPPHTTPLIB_OPENSSL_SUPPORT)
-    target_link_libraries(httplib INTERFACE ${OPENSSL_LIBRARIES})
-    if (WIN32)
-        target_link_libraries(httplib INTERFACE crypt32 cryptui ws2_32)
-    endif()
-    add_library(httplib::httplib ALIAS httplib)
+    set(HTTPLIB_REQUIRE_OPENSSL ON)
+    add_subdirectory(cpp-httplib EXCLUDE_FROM_ALL)
 endif()
 
 # cpp-jwt
 if (ENABLE_WEB_SERVICE AND NOT TARGET cpp-jwt::cpp-jwt)
-    add_library(cpp-jwt INTERFACE)
-    target_include_directories(cpp-jwt INTERFACE ./cpp-jwt/include)
-    target_compile_definitions(cpp-jwt INTERFACE CPP_JWT_USE_VENDORED_NLOHMANN_JSON)
-    add_library(cpp-jwt::cpp-jwt ALIAS cpp-jwt)
+    set(CPP_JWT_BUILD_EXAMPLES OFF)
+    set(CPP_JWT_BUILD_TESTS OFF)
+    set(CPP_JWT_USE_VENDORED_NLOHMANN_JSON OFF)
+    add_subdirectory(cpp-jwt EXCLUDE_FROM_ALL)
 endif()
 
 # Opus
@@ -155,3 +132,14 @@ endif()
 if (YUZU_USE_EXTERNAL_VULKAN_HEADERS)
     add_subdirectory(Vulkan-Headers EXCLUDE_FROM_ALL)
 endif()
+
+if (NOT TARGET LLVM::Demangle)
+    add_library(demangle STATIC)
+    target_include_directories(demangle PUBLIC ./demangle)
+    target_sources(demangle PRIVATE demangle/ItaniumDemangle.cpp)
+    add_library(LLVM::Demangle ALIAS demangle)
+endif()
+
+add_library(stb STATIC)
+target_include_directories(stb PUBLIC ./stb)
+target_sources(stb PRIVATE stb/stb_dxt.cpp)

externals/demangle/ItaniumDemangle.cpp

@@ -0,0 +1,588 @@
+//===------------------------- ItaniumDemangle.cpp ------------------------===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-FileCopyrightText: Part of the LLVM Project
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+
+// FIXME: (possibly) incomplete list of features that clang mangles that this
+// file does not yet support:
+//   - C++ modules TS
+
+#include "llvm/Demangle/Demangle.h"
+#include "llvm/Demangle/ItaniumDemangle.h"
+
+#include <cassert>
+#include <cctype>
+#include <cstdio>
+#include <cstdlib>
+#include <cstring>
+#include <functional>
+#include <numeric>
+#include <utility>
+#include <vector>
+
+using namespace llvm;
+using namespace llvm::itanium_demangle;
+
+constexpr const char *itanium_demangle::FloatData<float>::spec;
+constexpr const char *itanium_demangle::FloatData<double>::spec;
+constexpr const char *itanium_demangle::FloatData<long double>::spec;
+
+// <discriminator> := _ <non-negative number>      # when number < 10
+//                 := __ <non-negative number> _   # when number >= 10
+//  extension      := decimal-digit+               # at the end of string
+const char *itanium_demangle::parse_discriminator(const char *first,
+                                                  const char *last) {
+  // parse but ignore discriminator
+  if (first != last) {
+    if (*first == '_') {
+      const char *t1 = first + 1;
+      if (t1 != last) {
+        if (std::isdigit(*t1))
+          first = t1 + 1;
+        else if (*t1 == '_') {
+          for (++t1; t1 != last && std::isdigit(*t1); ++t1)
+            ;
+          if (t1 != last && *t1 == '_')
+            first = t1 + 1;
+        }
+      }
+    } else if (std::isdigit(*first)) {
+      const char *t1 = first + 1;
+      for (; t1 != last && std::isdigit(*t1); ++t1)
+        ;
+      if (t1 == last)
+        first = last;
+    }
+  }
+  return first;
+}
+
+#ifndef NDEBUG
+namespace {
+struct DumpVisitor {
+  unsigned Depth = 0;
+  bool PendingNewline = false;
+
+  template<typename NodeT> static constexpr bool wantsNewline(const NodeT *) {
+    return true;
+  }
+  static bool wantsNewline(NodeArray A) { return !A.empty(); }
+  static constexpr bool wantsNewline(...) { return false; }
+
+  template<typename ...Ts> static bool anyWantNewline(Ts ...Vs) {
+    for (bool B : {wantsNewline(Vs)...})
+      if (B)
+        return true;
+    return false;
+  }
+
+  void printStr(const char *S) { fprintf(stderr, "%s", S); }
+  void print(StringView SV) {
+    fprintf(stderr, "\"%.*s\"", (int)SV.size(), SV.begin());
+  }
+  void print(const Node *N) {
+    if (N)
+      N->visit(std::ref(*this));
+    else
+      printStr("<null>");
+  }
+  void print(NodeOrString NS) {
+    if (NS.isNode())
+      print(NS.asNode());
+    else if (NS.isString())
+      print(NS.asString());
+    else
+      printStr("NodeOrString()");
+  }
+  void print(NodeArray A) {
+    ++Depth;
+    printStr("{");
+    bool First = true;
+    for (const Node *N : A) {
+      if (First)
+        print(N);
+      else
+        printWithComma(N);
+      First = false;
+    }
+    printStr("}");
+    --Depth;
+  }
+
+  // Overload used when T is exactly 'bool', not merely convertible to 'bool'.
+  void print(bool B) { printStr(B ? "true" : "false"); }
+
+  template <class T>
+  typename std::enable_if<std::is_unsigned<T>::value>::type print(T N) {
+    fprintf(stderr, "%llu", (unsigned long long)N);
+  }
+
+  template <class T>
+  typename std::enable_if<std::is_signed<T>::value>::type print(T N) {
+    fprintf(stderr, "%lld", (long long)N);
+  }
+
+  void print(ReferenceKind RK) {
+    switch (RK) {
+    case ReferenceKind::LValue:
+      return printStr("ReferenceKind::LValue");
+    case ReferenceKind::RValue:
+      return printStr("ReferenceKind::RValue");
+    }
+  }
+  void print(FunctionRefQual RQ) {
+    switch (RQ) {
+    case FunctionRefQual::FrefQualNone:
+      return printStr("FunctionRefQual::FrefQualNone");
+    case FunctionRefQual::FrefQualLValue:
+      return printStr("FunctionRefQual::FrefQualLValue");
+    case FunctionRefQual::FrefQualRValue:
+      return printStr("FunctionRefQual::FrefQualRValue");
+    }
+  }
+  void print(Qualifiers Qs) {
+    if (!Qs) return printStr("QualNone");
+    struct QualName { Qualifiers Q; const char *Name; } Names[] = {
+      {QualConst, "QualConst"},
+      {QualVolatile, "QualVolatile"},
+      {QualRestrict, "QualRestrict"},
+    };
+    for (QualName Name : Names) {
+      if (Qs & Name.Q) {
+        printStr(Name.Name);
+        Qs = Qualifiers(Qs & ~Name.Q);
+        if (Qs) printStr(" | ");
+      }
+    }
+  }
+  void print(SpecialSubKind SSK) {
+    switch (SSK) {
+    case SpecialSubKind::allocator:
+      return printStr("SpecialSubKind::allocator");
+    case SpecialSubKind::basic_string:
+      return printStr("SpecialSubKind::basic_string");
+    case SpecialSubKind::string:
+      return printStr("SpecialSubKind::string");
+    case SpecialSubKind::istream:
+      return printStr("SpecialSubKind::istream");
+    case SpecialSubKind::ostream:
+      return printStr("SpecialSubKind::ostream");
+    case SpecialSubKind::iostream:
+      return printStr("SpecialSubKind::iostream");
+    }
+  }
+  void print(TemplateParamKind TPK) {
+    switch (TPK) {
+    case TemplateParamKind::Type:
+      return printStr("TemplateParamKind::Type");
+    case TemplateParamKind::NonType:
+      return printStr("TemplateParamKind::NonType");
+    case TemplateParamKind::Template:
+      return printStr("TemplateParamKind::Template");
+    }
+  }
+
+  void newLine() {
+    printStr("\n");
+    for (unsigned I = 0; I != Depth; ++I)
+      printStr(" ");
+    PendingNewline = false;
+  }
+
+  template<typename T> void printWithPendingNewline(T V) {
+    print(V);
+    if (wantsNewline(V))
+      PendingNewline = true;
+  }
+
+  template<typename T> void printWithComma(T V) {
+    if (PendingNewline || wantsNewline(V)) {
+      printStr(",");
+      newLine();
+    } else {
+      printStr(", ");
+    }
+
+    printWithPendingNewline(V);
+  }
+
+  struct CtorArgPrinter {
+    DumpVisitor &Visitor;
+
+    template<typename T, typename ...Rest> void operator()(T V, Rest ...Vs) {
+      if (Visitor.anyWantNewline(V, Vs...))
+        Visitor.newLine();
+      Visitor.printWithPendingNewline(V);
+      int PrintInOrder[] = { (Visitor.printWithComma(Vs), 0)..., 0 };
+      (void)PrintInOrder;
+    }
+  };
+
+  template<typename NodeT> void operator()(const NodeT *Node) {
+    Depth += 2;
+    fprintf(stderr, "%s(", itanium_demangle::NodeKind<NodeT>::name());
+    Node->match(CtorArgPrinter{*this});
+    fprintf(stderr, ")");
+    Depth -= 2;
+  }
+
+  void operator()(const ForwardTemplateReference *Node) {
+    Depth += 2;
+    fprintf(stderr, "ForwardTemplateReference(");
+    if (Node->Ref && !Node->Printing) {
+      Node->Printing = true;
+      CtorArgPrinter{*this}(Node->Ref);
+      Node->Printing = false;
+    } else {
+      CtorArgPrinter{*this}(Node->Index);
+    }
+    fprintf(stderr, ")");
+    Depth -= 2;
+  }
+};
+}
+
+void itanium_demangle::Node::dump() const {
+  DumpVisitor V;
+  visit(std::ref(V));
+  V.newLine();
+}
+#endif
+
+namespace {
+class BumpPointerAllocator {
+  struct BlockMeta {
+    BlockMeta* Next;
+    size_t Current;
+  };
+
+  static constexpr size_t AllocSize = 4096;
+  static constexpr size_t UsableAllocSize = AllocSize - sizeof(BlockMeta);
+
+  alignas(long double) char InitialBuffer[AllocSize];
+  BlockMeta* BlockList = nullptr;
+
+  void grow() {
+    char* NewMeta = static_cast<char *>(std::malloc(AllocSize));
+    if (NewMeta == nullptr)
+      std::terminate();
+    BlockList = new (NewMeta) BlockMeta{BlockList, 0};
+  }
+
+  void* allocateMassive(size_t NBytes) {
+    NBytes += sizeof(BlockMeta);
+    BlockMeta* NewMeta = reinterpret_cast<BlockMeta*>(std::malloc(NBytes));
+    if (NewMeta == nullptr)
+      std::terminate();
+    BlockList->Next = new (NewMeta) BlockMeta{BlockList->Next, 0};
+    return static_cast<void*>(NewMeta + 1);
+  }
+
+public:
+  BumpPointerAllocator()
+      : BlockList(new (InitialBuffer) BlockMeta{nullptr, 0}) {}
+
+  void* allocate(size_t N) {
+    N = (N + 15u) & ~15u;
+    if (N + BlockList->Current >= UsableAllocSize) {
+      if (N > UsableAllocSize)
+        return allocateMassive(N);
+      grow();
+    }
+    BlockList->Current += N;
+    return static_cast<void*>(reinterpret_cast<char*>(BlockList + 1) +
+                              BlockList->Current - N);
+  }
+
+  void reset() {
+    while (BlockList) {
+      BlockMeta* Tmp = BlockList;
+      BlockList = BlockList->Next;
+      if (reinterpret_cast<char*>(Tmp) != InitialBuffer)
+        std::free(Tmp);
+    }
+    BlockList = new (InitialBuffer) BlockMeta{nullptr, 0};
+  }
+
+  ~BumpPointerAllocator() { reset(); }
+};
+
+class DefaultAllocator {
+  BumpPointerAllocator Alloc;
+
+public:
+  void reset() { Alloc.reset(); }
+
+  template<typename T, typename ...Args> T *makeNode(Args &&...args) {
+    return new (Alloc.allocate(sizeof(T)))
+        T(std::forward<Args>(args)...);
+  }
+
+  void *allocateNodeArray(size_t sz) {
+    return Alloc.allocate(sizeof(Node *) * sz);
+  }
+};
+}  // unnamed namespace
+
+//===----------------------------------------------------------------------===//
+// Code beyond this point should not be synchronized with libc++abi.
+//===----------------------------------------------------------------------===//
+
+using Demangler = itanium_demangle::ManglingParser<DefaultAllocator>;
+
+char *llvm::itaniumDemangle(const char *MangledName, char *Buf,
+                            size_t *N, int *Status) {
+  if (MangledName == nullptr || (Buf != nullptr && N == nullptr)) {
+    if (Status)
+      *Status = demangle_invalid_args;
+    return nullptr;
+  }
+
+  int InternalStatus = demangle_success;
+  Demangler Parser(MangledName, MangledName + std::strlen(MangledName));
+  OutputStream S;
+
+  Node *AST = Parser.parse();
+
+  if (AST == nullptr)
+    InternalStatus = demangle_invalid_mangled_name;
+  else if (!initializeOutputStream(Buf, N, S, 1024))
+    InternalStatus = demangle_memory_alloc_failure;
+  else {
+    assert(Parser.ForwardTemplateRefs.empty());
+    AST->print(S);
+    S += '\0';
+    if (N != nullptr)
+      *N = S.getCurrentPosition();
+    Buf = S.getBuffer();
+  }
+
+  if (Status)
+    *Status = InternalStatus;
+  return InternalStatus == demangle_success ? Buf : nullptr;
+}
+
+ItaniumPartialDemangler::ItaniumPartialDemangler()
+    : RootNode(nullptr), Context(new Demangler{nullptr, nullptr}) {}
+
+ItaniumPartialDemangler::~ItaniumPartialDemangler() {
+  delete static_cast<Demangler *>(Context);
+}
+
+ItaniumPartialDemangler::ItaniumPartialDemangler(
+    ItaniumPartialDemangler &&Other)
+    : RootNode(Other.RootNode), Context(Other.Context) {
+  Other.Context = Other.RootNode = nullptr;
+}
+
+ItaniumPartialDemangler &ItaniumPartialDemangler::
+operator=(ItaniumPartialDemangler &&Other) {
+  std::swap(RootNode, Other.RootNode);
+  std::swap(Context, Other.Context);
+  return *this;
+}
+
+// Demangle MangledName into an AST, storing it into this->RootNode.
+bool ItaniumPartialDemangler::partialDemangle(const char *MangledName) {
+  Demangler *Parser = static_cast<Demangler *>(Context);
+  size_t Len = std::strlen(MangledName);
+  Parser->reset(MangledName, MangledName + Len);
+  RootNode = Parser->parse();
+  return RootNode == nullptr;
+}
+
+static char *printNode(const Node *RootNode, char *Buf, size_t *N) {
+  OutputStream S;
+  if (!initializeOutputStream(Buf, N, S, 128))
+    return nullptr;
+  RootNode->print(S);
+  S += '\0';
+  if (N != nullptr)
+    *N = S.getCurrentPosition();
+  return S.getBuffer();
+}
+
+char *ItaniumPartialDemangler::getFunctionBaseName(char *Buf, size_t *N) const {
+  if (!isFunction())
+    return nullptr;
+
+  const Node *Name = static_cast<const FunctionEncoding *>(RootNode)->getName();
+
+  while (true) {
+    switch (Name->getKind()) {
+    case Node::KAbiTagAttr:
+      Name = static_cast<const AbiTagAttr *>(Name)->Base;
+      continue;
+    case Node::KStdQualifiedName:
+      Name = static_cast<const StdQualifiedName *>(Name)->Child;
+      continue;
+    case Node::KNestedName:
+      Name = static_cast<const NestedName *>(Name)->Name;
+      continue;
+    case Node::KLocalName:
+      Name = static_cast<const LocalName *>(Name)->Entity;
+      continue;
+    case Node::KNameWithTemplateArgs:
+      Name = static_cast<const NameWithTemplateArgs *>(Name)->Name;
+      continue;
+    default:
+      return printNode(Name, Buf, N);
+    }
+  }
+}
+
+char *ItaniumPartialDemangler::getFunctionDeclContextName(char *Buf,
+                                                          size_t *N) const {
+  if (!isFunction())
+    return nullptr;
+  const Node *Name = static_cast<const FunctionEncoding *>(RootNode)->getName();
+
+  OutputStream S;
+  if (!initializeOutputStream(Buf, N, S, 128))
+    return nullptr;
+
+ KeepGoingLocalFunction:
+  while (true) {
+    if (Name->getKind() == Node::KAbiTagAttr) {
+      Name = static_cast<const AbiTagAttr *>(Name)->Base;
+      continue;
+    }
+    if (Name->getKind() == Node::KNameWithTemplateArgs) {
+      Name = static_cast<const NameWithTemplateArgs *>(Name)->Name;
+      continue;
+    }
+    break;
+  }
+
+  switch (Name->getKind()) {
+  case Node::KStdQualifiedName:
+    S += "std";
+    break;
+  case Node::KNestedName:
+    static_cast<const NestedName *>(Name)->Qual->print(S);
+    break;
+  case Node::KLocalName: {
+    auto *LN = static_cast<const LocalName *>(Name);
+    LN->Encoding->print(S);
+    S += "::";
+    Name = LN->Entity;
+    goto KeepGoingLocalFunction;
+  }
+  default:
+    break;
+  }
+  S += '\0';
+  if (N != nullptr)
+    *N = S.getCurrentPosition();
+  return S.getBuffer();
+}
+
+char *ItaniumPartialDemangler::getFunctionName(char *Buf, size_t *N) const {
+  if (!isFunction())
+    return nullptr;
+  auto *Name = static_cast<FunctionEncoding *>(RootNode)->getName();
+  return printNode(Name, Buf, N);
+}
+
+char *ItaniumPartialDemangler::getFunctionParameters(char *Buf,
+                                                     size_t *N) const {
+  if (!isFunction())
+    return nullptr;
+  NodeArray Params = static_cast<FunctionEncoding *>(RootNode)->getParams();
+
+  OutputStream S;
+  if (!initializeOutputStream(Buf, N, S, 128))
+    return nullptr;
+
+  S += '(';
+  Params.printWithComma(S);
+  S += ')';
+  S += '\0';
+  if (N != nullptr)
+    *N = S.getCurrentPosition();
+  return S.getBuffer();
+}
+
+char *ItaniumPartialDemangler::getFunctionReturnType(
+    char *Buf, size_t *N) const {
+  if (!isFunction())
+    return nullptr;
+
+  OutputStream S;
+  if (!initializeOutputStream(Buf, N, S, 128))
+    return nullptr;
+
+  if (const Node *Ret =
+          static_cast<const FunctionEncoding *>(RootNode)->getReturnType())
+    Ret->print(S);
+
+  S += '\0';
+  if (N != nullptr)
+    *N = S.getCurrentPosition();
+  return S.getBuffer();
+}
+
+char *ItaniumPartialDemangler::finishDemangle(char *Buf, size_t *N) const {
+  assert(RootNode != nullptr && "must call partialDemangle()");
+  return printNode(static_cast<Node *>(RootNode), Buf, N);
+}
+
+bool ItaniumPartialDemangler::hasFunctionQualifiers() const {
+  assert(RootNode != nullptr && "must call partialDemangle()");
+  if (!isFunction())
+    return false;
+  auto *E = static_cast<const FunctionEncoding *>(RootNode);
+  return E->getCVQuals() != QualNone || E->getRefQual() != FrefQualNone;
+}
+
+bool ItaniumPartialDemangler::isCtorOrDtor() const {
+  const Node *N = static_cast<const Node *>(RootNode);
+  while (N) {
+    switch (N->getKind()) {
+    default:
+      return false;
+    case Node::KCtorDtorName:
+      return true;
+
+    case Node::KAbiTagAttr:
+      N = static_cast<const AbiTagAttr *>(N)->Base;
+      break;
+    case Node::KFunctionEncoding:
+      N = static_cast<const FunctionEncoding *>(N)->getName();
+      break;
+    case Node::KLocalName:
+      N = static_cast<const LocalName *>(N)->Entity;
+      break;
+    case Node::KNameWithTemplateArgs:
+      N = static_cast<const NameWithTemplateArgs *>(N)->Name;
+      break;
+    case Node::KNestedName:
+      N = static_cast<const NestedName *>(N)->Name;
+      break;
+    case Node::KStdQualifiedName:
+      N = static_cast<const StdQualifiedName *>(N)->Child;
+      break;
+    }
+  }
+  return false;
+}
+
+bool ItaniumPartialDemangler::isFunction() const {
+  assert(RootNode != nullptr && "must call partialDemangle()");
+  return static_cast<const Node *>(RootNode)->getKind() ==
+         Node::KFunctionEncoding;
+}
+
+bool ItaniumPartialDemangler::isSpecialName() const {
+  assert(RootNode != nullptr && "must call partialDemangle()");
+  auto K = static_cast<const Node *>(RootNode)->getKind();
+  return K == Node::KSpecialName || K == Node::KCtorVtableSpecialName;
+}
+
+bool ItaniumPartialDemangler::isData() const {
+  return !isFunction() && !isSpecialName();
+}

externals/demangle/llvm/Demangle/Demangle.h

@@ -0,0 +1,104 @@
+//===--- Demangle.h ---------------------------------------------*- C++ -*-===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-FileCopyrightText: Part of the LLVM Project
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+
+#ifndef LLVM_DEMANGLE_DEMANGLE_H
+#define LLVM_DEMANGLE_DEMANGLE_H
+
+#include <cstddef>
+#include <string>
+
+namespace llvm {
+/// This is a llvm local version of __cxa_demangle. Other than the name and
+/// being in the llvm namespace it is identical.
+///
+/// The mangled_name is demangled into buf and returned. If the buffer is not
+/// large enough, realloc is used to expand it.
+///
+/// The *status will be set to a value from the following enumeration
+enum : int {
+  demangle_unknown_error = -4,
+  demangle_invalid_args = -3,
+  demangle_invalid_mangled_name = -2,
+  demangle_memory_alloc_failure = -1,
+  demangle_success = 0,
+};
+
+char *itaniumDemangle(const char *mangled_name, char *buf, size_t *n,
+                      int *status);
+
+
+enum MSDemangleFlags {
+  MSDF_None = 0,
+  MSDF_DumpBackrefs = 1 << 0,
+  MSDF_NoAccessSpecifier = 1 << 1,
+  MSDF_NoCallingConvention = 1 << 2,
+  MSDF_NoReturnType = 1 << 3,
+  MSDF_NoMemberType = 1 << 4,
+};
+char *microsoftDemangle(const char *mangled_name, char *buf, size_t *n,
+                        int *status, MSDemangleFlags Flags = MSDF_None);
+
+/// "Partial" demangler. This supports demangling a string into an AST
+/// (typically an intermediate stage in itaniumDemangle) and querying certain
+/// properties or partially printing the demangled name.
+struct ItaniumPartialDemangler {
+  ItaniumPartialDemangler();
+
+  ItaniumPartialDemangler(ItaniumPartialDemangler &&Other);
+  ItaniumPartialDemangler &operator=(ItaniumPartialDemangler &&Other);
+
+  /// Demangle into an AST. Subsequent calls to the rest of the member functions
+  /// implicitly operate on the AST this produces.
+  /// \return true on error, false otherwise
+  bool partialDemangle(const char *MangledName);
+
+  /// Just print the entire mangled name into Buf. Buf and N behave like the
+  /// second and third parameters to itaniumDemangle.
+  char *finishDemangle(char *Buf, size_t *N) const;
+
+  /// Get the base name of a function. This doesn't include trailing template
+  /// arguments, ie for "a::b<int>" this function returns "b".
+  char *getFunctionBaseName(char *Buf, size_t *N) const;
+
+  /// Get the context name for a function. For "a::b::c", this function returns
+  /// "a::b".
+  char *getFunctionDeclContextName(char *Buf, size_t *N) const;
+
+  /// Get the entire name of this function.
+  char *getFunctionName(char *Buf, size_t *N) const;
+
+  /// Get the parameters for this function.
+  char *getFunctionParameters(char *Buf, size_t *N) const;
+  char *getFunctionReturnType(char *Buf, size_t *N) const;
+
+  /// If this function has any any cv or reference qualifiers. These imply that
+  /// the function is a non-static member function.
+  bool hasFunctionQualifiers() const;
+
+  /// If this symbol describes a constructor or destructor.
+  bool isCtorOrDtor() const;
+
+  /// If this symbol describes a function.
+  bool isFunction() const;
+
+  /// If this symbol describes a variable.
+  bool isData() const;
+
+  /// If this symbol is a <special-name>. These are generally implicitly
+  /// generated by the implementation, such as vtables and typeinfo names.
+  bool isSpecialName() const;
+
+  ~ItaniumPartialDemangler();
+private:
+  void *RootNode;
+  void *Context;
+};
+} // namespace llvm
+
+#endif

externals/demangle/llvm/Demangle/DemangleConfig.h

@@ -0,0 +1,93 @@
+//===--- DemangleConfig.h ---------------------------------------*- C++ -*-===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-FileCopyrightText: Part of the LLVM Project
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+//
+// This file contains a variety of feature test macros copied from
+// include/llvm/Support/Compiler.h so that LLVMDemangle does not need to take
+// a dependency on LLVMSupport.
+//
+//===----------------------------------------------------------------------===//
+
+#ifndef LLVM_DEMANGLE_COMPILER_H
+#define LLVM_DEMANGLE_COMPILER_H
+
+#ifndef __has_feature
+#define __has_feature(x) 0
+#endif
+
+#ifndef __has_cpp_attribute
+#define __has_cpp_attribute(x) 0
+#endif
+
+#ifndef __has_attribute
+#define __has_attribute(x) 0
+#endif
+
+#ifndef __has_builtin
+#define __has_builtin(x) 0
+#endif
+
+#ifndef DEMANGLE_GNUC_PREREQ
+#if defined(__GNUC__) && defined(__GNUC_MINOR__) && defined(__GNUC_PATCHLEVEL__)
+#define DEMANGLE_GNUC_PREREQ(maj, min, patch)                           \
+  ((__GNUC__ << 20) + (__GNUC_MINOR__ << 10) + __GNUC_PATCHLEVEL__ >=          \
+   ((maj) << 20) + ((min) << 10) + (patch))
+#elif defined(__GNUC__) && defined(__GNUC_MINOR__)
+#define DEMANGLE_GNUC_PREREQ(maj, min, patch)                           \
+  ((__GNUC__ << 20) + (__GNUC_MINOR__ << 10) >= ((maj) << 20) + ((min) << 10))
+#else
+#define DEMANGLE_GNUC_PREREQ(maj, min, patch) 0
+#endif
+#endif
+
+#if __has_attribute(used) || DEMANGLE_GNUC_PREREQ(3, 1, 0)
+#define DEMANGLE_ATTRIBUTE_USED __attribute__((__used__))
+#else
+#define DEMANGLE_ATTRIBUTE_USED
+#endif
+
+#if __has_builtin(__builtin_unreachable) || DEMANGLE_GNUC_PREREQ(4, 5, 0)
+#define DEMANGLE_UNREACHABLE __builtin_unreachable()
+#elif defined(_MSC_VER)
+#define DEMANGLE_UNREACHABLE __assume(false)
+#else
+#define DEMANGLE_UNREACHABLE
+#endif
+
+#if __has_attribute(noinline) || DEMANGLE_GNUC_PREREQ(3, 4, 0)
+#define DEMANGLE_ATTRIBUTE_NOINLINE __attribute__((noinline))
+#elif defined(_MSC_VER)
+#define DEMANGLE_ATTRIBUTE_NOINLINE __declspec(noinline)
+#else
+#define DEMANGLE_ATTRIBUTE_NOINLINE
+#endif
+
+#if !defined(NDEBUG)
+#define DEMANGLE_DUMP_METHOD DEMANGLE_ATTRIBUTE_NOINLINE DEMANGLE_ATTRIBUTE_USED
+#else
+#define DEMANGLE_DUMP_METHOD DEMANGLE_ATTRIBUTE_NOINLINE
+#endif
+
+#if __cplusplus > 201402L && __has_cpp_attribute(fallthrough)
+#define DEMANGLE_FALLTHROUGH [[fallthrough]]
+#elif __has_cpp_attribute(gnu::fallthrough)
+#define DEMANGLE_FALLTHROUGH [[gnu::fallthrough]]
+#elif !__cplusplus
+// Workaround for llvm.org/PR23435, since clang 3.6 and below emit a spurious
+// error when __has_cpp_attribute is given a scoped attribute in C mode.
+#define DEMANGLE_FALLTHROUGH
+#elif __has_cpp_attribute(clang::fallthrough)
+#define DEMANGLE_FALLTHROUGH [[clang::fallthrough]]
+#else
+#define DEMANGLE_FALLTHROUGH
+#endif
+
+#define DEMANGLE_NAMESPACE_BEGIN namespace llvm { namespace itanium_demangle {
+#define DEMANGLE_NAMESPACE_END } }
+
+#endif

externals/demangle/llvm/Demangle/StringView.h

@@ -0,0 +1,127 @@
+//===--- StringView.h -------------------------------------------*- C++ -*-===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-FileCopyrightText: Part of the LLVM Project
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+//
+// FIXME: Use std::string_view instead when we support C++17.
+//
+//===----------------------------------------------------------------------===//
+
+#ifndef DEMANGLE_STRINGVIEW_H
+#define DEMANGLE_STRINGVIEW_H
+
+#include "DemangleConfig.h"
+#include <algorithm>
+#include <cassert>
+#include <cstring>
+
+DEMANGLE_NAMESPACE_BEGIN
+
+class StringView {
+  const char *First;
+  const char *Last;
+
+public:
+  static const size_t npos = ~size_t(0);
+
+  template <size_t N>
+  StringView(const char (&Str)[N]) : First(Str), Last(Str + N - 1) {}
+  StringView(const char *First_, const char *Last_)
+      : First(First_), Last(Last_) {}
+  StringView(const char *First_, size_t Len)
+      : First(First_), Last(First_ + Len) {}
+  StringView(const char *Str) : First(Str), Last(Str + std::strlen(Str)) {}
+  StringView() : First(nullptr), Last(nullptr) {}
+
+  StringView substr(size_t From) const {
+    return StringView(begin() + From, size() - From);
+  }
+
+  size_t find(char C, size_t From = 0) const {
+    size_t FindBegin = std::min(From, size());
+    // Avoid calling memchr with nullptr.
+    if (FindBegin < size()) {
+      // Just forward to memchr, which is faster than a hand-rolled loop.
+      if (const void *P = ::memchr(First + FindBegin, C, size() - FindBegin))
+        return size_t(static_cast<const char *>(P) - First);
+    }
+    return npos;
+  }
+
+  StringView substr(size_t From, size_t To) const {
+    if (To >= size())
+      To = size() - 1;
+    if (From >= size())
+      From = size() - 1;
+    return StringView(First + From, First + To);
+  }
+
+  StringView dropFront(size_t N = 1) const {
+    if (N >= size())
+      N = size();
+    return StringView(First + N, Last);
+  }
+
+  StringView dropBack(size_t N = 1) const {
+    if (N >= size())
+      N = size();
+    return StringView(First, Last - N);
+  }
+
+  char front() const {
+    assert(!empty());
+    return *begin();
+  }
+
+  char back() const {
+    assert(!empty());
+    return *(end() - 1);
+  }
+
+  char popFront() {
+    assert(!empty());
+    return *First++;
+  }
+
+  bool consumeFront(char C) {
+    if (!startsWith(C))
+      return false;
+    *this = dropFront(1);
+    return true;
+  }
+
+  bool consumeFront(StringView S) {
+    if (!startsWith(S))
+      return false;
+    *this = dropFront(S.size());
+    return true;
+  }
+
+  bool startsWith(char C) const { return !empty() && *begin() == C; }
+
+  bool startsWith(StringView Str) const {
+    if (Str.size() > size())
+      return false;
+    return std::equal(Str.begin(), Str.end(), begin());
+  }
+
+  const char &operator[](size_t Idx) const { return *(begin() + Idx); }
+
+  const char *begin() const { return First; }
+  const char *end() const { return Last; }
+  size_t size() const { return static_cast<size_t>(Last - First); }
+  bool empty() const { return First == Last; }
+};
+
+inline bool operator==(const StringView &LHS, const StringView &RHS) {
+  return LHS.size() == RHS.size() &&
+         std::equal(LHS.begin(), LHS.end(), RHS.begin());
+}
+
+DEMANGLE_NAMESPACE_END
+
+#endif

externals/demangle/llvm/Demangle/Utility.h

@@ -0,0 +1,192 @@
+//===--- Utility.h ----------------------------------------------*- C++ -*-===//
+//
+// Part of the LLVM Project, under the Apache License v2.0 with LLVM Exceptions.
+// See https://llvm.org/LICENSE.txt for license information.
+// SPDX-FileCopyrightText: Part of the LLVM Project
+// SPDX-License-Identifier: Apache-2.0 WITH LLVM-exception
+//
+//===----------------------------------------------------------------------===//
+//
+// Provide some utility classes for use in the demangler(s).
+//
+//===----------------------------------------------------------------------===//
+
+#ifndef DEMANGLE_UTILITY_H
+#define DEMANGLE_UTILITY_H
+
+#include "StringView.h"
+#include <cstdint>
+#include <cstdlib>
+#include <cstring>
+#include <iterator>
+#include <limits>
+
+DEMANGLE_NAMESPACE_BEGIN
+
+// Stream that AST nodes write their string representation into after the AST
+// has been parsed.
+class OutputStream {
+  char *Buffer;
+  size_t CurrentPosition;
+  size_t BufferCapacity;
+
+  // Ensure there is at least n more positions in buffer.
+  void grow(size_t N) {
+    if (N + CurrentPosition >= BufferCapacity) {
+      BufferCapacity *= 2;
+      if (BufferCapacity < N + CurrentPosition)
+        BufferCapacity = N + CurrentPosition;
+      Buffer = static_cast<char *>(std::realloc(Buffer, BufferCapacity));
+      if (Buffer == nullptr)
+        std::terminate();
+    }
+  }
+
+  void writeUnsigned(uint64_t N, bool isNeg = false) {
+    // Handle special case...
+    if (N == 0) {
+      *this << '0';
+      return;
+    }
+
+    char Temp[21];
+    char *TempPtr = std::end(Temp);
+
+    while (N) {
+      *--TempPtr = '0' + char(N % 10);
+      N /= 10;
+    }
+
+    // Add negative sign...
+    if (isNeg)
+      *--TempPtr = '-';
+    this->operator<<(StringView(TempPtr, std::end(Temp)));
+  }
+
+public:
+  OutputStream(char *StartBuf, size_t Size)
+      : Buffer(StartBuf), CurrentPosition(0), BufferCapacity(Size) {}
+  OutputStream() = default;
+  void reset(char *Buffer_, size_t BufferCapacity_) {
+    CurrentPosition = 0;
+    Buffer = Buffer_;
+    BufferCapacity = BufferCapacity_;
+  }
+
+  /// If a ParameterPackExpansion (or similar type) is encountered, the offset
+  /// into the pack that we're currently printing.
+  unsigned CurrentPackIndex = std::numeric_limits<unsigned>::max();
+  unsigned CurrentPackMax = std::numeric_limits<unsigned>::max();
+
+  OutputStream &operator+=(StringView R) {
+    size_t Size = R.size();
+    if (Size == 0)
+      return *this;
+    grow(Size);
+    std::memmove(Buffer + CurrentPosition, R.begin(), Size);
+    CurrentPosition += Size;
+    return *this;
+  }
+
+  OutputStream &operator+=(char C) {
+    grow(1);
+    Buffer[CurrentPosition++] = C;
+    return *this;
+  }
+
+  OutputStream &operator<<(StringView R) { return (*this += R); }
+
+  OutputStream &operator<<(char C) { return (*this += C); }
+
+  OutputStream &operator<<(long long N) {
+    if (N < 0)
+      writeUnsigned(static_cast<unsigned long long>(-N), true);
+    else
+      writeUnsigned(static_cast<unsigned long long>(N));
+    return *this;
+  }
+
+  OutputStream &operator<<(unsigned long long N) {
+    writeUnsigned(N, false);
+    return *this;
+  }
+
+  OutputStream &operator<<(long N) {
+    return this->operator<<(static_cast<long long>(N));
+  }
+
+  OutputStream &operator<<(unsigned long N) {
+    return this->operator<<(static_cast<unsigned long long>(N));
+  }
+
+  OutputStream &operator<<(int N) {
+    return this->operator<<(static_cast<long long>(N));
+  }
+
+  OutputStream &operator<<(unsigned int N) {
+    return this->operator<<(static_cast<unsigned long long>(N));
+  }
+
+  size_t getCurrentPosition() const { return CurrentPosition; }
+  void setCurrentPosition(size_t NewPos) { CurrentPosition = NewPos; }
+
+  char back() const {
+    return CurrentPosition ? Buffer[CurrentPosition - 1] : '\0';
+  }
+
+  bool empty() const { return CurrentPosition == 0; }
+
+  char *getBuffer() { return Buffer; }
+  char *getBufferEnd() { return Buffer + CurrentPosition - 1; }
+  size_t getBufferCapacity() { return BufferCapacity; }
+};
+
+template <class T> class SwapAndRestore {
+  T &Restore;
+  T OriginalValue;
+  bool ShouldRestore = true;
+
+public:
+  SwapAndRestore(T &Restore_) : SwapAndRestore(Restore_, Restore_) {}
+
+  SwapAndRestore(T &Restore_, T NewVal)
+      : Restore(Restore_), OriginalValue(Restore) {
+    Restore = std::move(NewVal);
+  }
+  ~SwapAndRestore() {
+    if (ShouldRestore)
+      Restore = std::move(OriginalValue);
+  }
+
+  void shouldRestore(bool ShouldRestore_) { ShouldRestore = ShouldRestore_; }
+
+  void restoreNow(bool Force) {
+    if (!Force && !ShouldRestore)
+      return;
+
+    Restore = std::move(OriginalValue);
+    ShouldRestore = false;
+  }
+
+  SwapAndRestore(const SwapAndRestore &) = delete;
+  SwapAndRestore &operator=(const SwapAndRestore &) = delete;
+};
+
+inline bool initializeOutputStream(char *Buf, size_t *N, OutputStream &S,
+                                   size_t InitSize) {
+  size_t BufferSize;
+  if (Buf == nullptr) {
+    Buf = static_cast<char *>(std::malloc(InitSize));
+    if (Buf == nullptr)
+      return false;
+    BufferSize = InitSize;
+  } else
+    BufferSize = *N;
+
+  S.reset(Buf, BufferSize);
+  return true;
+}
+
+DEMANGLE_NAMESPACE_END
+
+#endif

externals/find-modules/Findinih.cmake

@@ -1,16 +0,0 @@
-# SPDX-FileCopyrightText: 2022 Alexandre Bouvier <contact@amb.tf>
-#
-# SPDX-License-Identifier: GPL-3.0-or-later
-
-find_package(PkgConfig QUIET)
-pkg_search_module(INIREADER QUIET IMPORTED_TARGET INIReader)
-
-include(FindPackageHandleStandardArgs)
-find_package_handle_standard_args(inih
-    REQUIRED_VARS INIREADER_LINK_LIBRARIES
-    VERSION_VAR INIREADER_VERSION
-)
-
-if (inih_FOUND AND NOT TARGET inih::INIReader)
-    add_library(inih::INIReader ALIAS PkgConfig::INIREADER)
-endif()

externals/stb/stb_dxt.cpp

@@ -0,0 +1,765 @@
+// SPDX-FileCopyrightText: fabian "ryg" giesen
+// SPDX-License-Identifier: MIT
+
+// stb_dxt.h - v1.12 - DXT1/DXT5 compressor
+
+#include <stb_dxt.h>
+
+#include <stdlib.h>
+#include <string.h>
+
+#if !defined(STBD_FABS)
+#include <math.h>
+#endif
+
+#ifndef STBD_FABS
+#define STBD_FABS(x) fabs(x)
+#endif
+
+static const unsigned char stb__OMatch5[256][2] = {
+    {0, 0},   {0, 0},   {0, 1},   {0, 1},   {1, 0},   {1, 0},   {1, 0},   {1, 1},   {1, 1},
+    {1, 1},   {1, 2},   {0, 4},   {2, 1},   {2, 1},   {2, 1},   {2, 2},   {2, 2},   {2, 2},
+    {2, 3},   {1, 5},   {3, 2},   {3, 2},   {4, 0},   {3, 3},   {3, 3},   {3, 3},   {3, 4},
+    {3, 4},   {3, 4},   {3, 5},   {4, 3},   {4, 3},   {5, 2},   {4, 4},   {4, 4},   {4, 5},
+    {4, 5},   {5, 4},   {5, 4},   {5, 4},   {6, 3},   {5, 5},   {5, 5},   {5, 6},   {4, 8},
+    {6, 5},   {6, 5},   {6, 5},   {6, 6},   {6, 6},   {6, 6},   {6, 7},   {5, 9},   {7, 6},
+    {7, 6},   {8, 4},   {7, 7},   {7, 7},   {7, 7},   {7, 8},   {7, 8},   {7, 8},   {7, 9},
+    {8, 7},   {8, 7},   {9, 6},   {8, 8},   {8, 8},   {8, 9},   {8, 9},   {9, 8},   {9, 8},
+    {9, 8},   {10, 7},  {9, 9},   {9, 9},   {9, 10},  {8, 12},  {10, 9},  {10, 9},  {10, 9},
+    {10, 10}, {10, 10}, {10, 10}, {10, 11}, {9, 13},  {11, 10}, {11, 10}, {12, 8},  {11, 11},
+    {11, 11}, {11, 11}, {11, 12}, {11, 12}, {11, 12}, {11, 13}, {12, 11}, {12, 11}, {13, 10},
+    {12, 12}, {12, 12}, {12, 13}, {12, 13}, {13, 12}, {13, 12}, {13, 12}, {14, 11}, {13, 13},
+    {13, 13}, {13, 14}, {12, 16}, {14, 13}, {14, 13}, {14, 13}, {14, 14}, {14, 14}, {14, 14},
+    {14, 15}, {13, 17}, {15, 14}, {15, 14}, {16, 12}, {15, 15}, {15, 15}, {15, 15}, {15, 16},
+    {15, 16}, {15, 16}, {15, 17}, {16, 15}, {16, 15}, {17, 14}, {16, 16}, {16, 16}, {16, 17},
+    {16, 17}, {17, 16}, {17, 16}, {17, 16}, {18, 15}, {17, 17}, {17, 17}, {17, 18}, {16, 20},
+    {18, 17}, {18, 17}, {18, 17}, {18, 18}, {18, 18}, {18, 18}, {18, 19}, {17, 21}, {19, 18},
+    {19, 18}, {20, 16}, {19, 19}, {19, 19}, {19, 19}, {19, 20}, {19, 20}, {19, 20}, {19, 21},
+    {20, 19}, {20, 19}, {21, 18}, {20, 20}, {20, 20}, {20, 21}, {20, 21}, {21, 20}, {21, 20},
+    {21, 20}, {22, 19}, {21, 21}, {21, 21}, {21, 22}, {20, 24}, {22, 21}, {22, 21}, {22, 21},
+    {22, 22}, {22, 22}, {22, 22}, {22, 23}, {21, 25}, {23, 22}, {23, 22}, {24, 20}, {23, 23},
+    {23, 23}, {23, 23}, {23, 24}, {23, 24}, {23, 24}, {23, 25}, {24, 23}, {24, 23}, {25, 22},
+    {24, 24}, {24, 24}, {24, 25}, {24, 25}, {25, 24}, {25, 24}, {25, 24}, {26, 23}, {25, 25},
+    {25, 25}, {25, 26}, {24, 28}, {26, 25}, {26, 25}, {26, 25}, {26, 26}, {26, 26}, {26, 26},
+    {26, 27}, {25, 29}, {27, 26}, {27, 26}, {28, 24}, {27, 27}, {27, 27}, {27, 27}, {27, 28},
+    {27, 28}, {27, 28}, {27, 29}, {28, 27}, {28, 27}, {29, 26}, {28, 28}, {28, 28}, {28, 29},
+    {28, 29}, {29, 28}, {29, 28}, {29, 28}, {30, 27}, {29, 29}, {29, 29}, {29, 30}, {29, 30},
+    {30, 29}, {30, 29}, {30, 29}, {30, 30}, {30, 30}, {30, 30}, {30, 31}, {30, 31}, {31, 30},
+    {31, 30}, {31, 30}, {31, 31}, {31, 31},
+};
+static const unsigned char stb__OMatch6[256][2] = {
+    {0, 0},   {0, 1},   {1, 0},   {1, 1},   {1, 1},   {1, 2},   {2, 1},   {2, 2},   {2, 2},
+    {2, 3},   {3, 2},   {3, 3},   {3, 3},   {3, 4},   {4, 3},   {4, 4},   {4, 4},   {4, 5},
+    {5, 4},   {5, 5},   {5, 5},   {5, 6},   {6, 5},   {6, 6},   {6, 6},   {6, 7},   {7, 6},
+    {7, 7},   {7, 7},   {7, 8},   {8, 7},   {8, 8},   {8, 8},   {8, 9},   {9, 8},   {9, 9},
+    {9, 9},   {9, 10},  {10, 9},  {10, 10}, {10, 10}, {10, 11}, {11, 10}, {8, 16},  {11, 11},
+    {11, 12}, {12, 11}, {9, 17},  {12, 12}, {12, 13}, {13, 12}, {11, 16}, {13, 13}, {13, 14},
+    {14, 13}, {12, 17}, {14, 14}, {14, 15}, {15, 14}, {14, 16}, {15, 15}, {15, 16}, {16, 14},
+    {16, 15}, {17, 14}, {16, 16}, {16, 17}, {17, 16}, {18, 15}, {17, 17}, {17, 18}, {18, 17},
+    {20, 14}, {18, 18}, {18, 19}, {19, 18}, {21, 15}, {19, 19}, {19, 20}, {20, 19}, {20, 20},
+    {20, 20}, {20, 21}, {21, 20}, {21, 21}, {21, 21}, {21, 22}, {22, 21}, {22, 22}, {22, 22},
+    {22, 23}, {23, 22}, {23, 23}, {23, 23}, {23, 24}, {24, 23}, {24, 24}, {24, 24}, {24, 25},
+    {25, 24}, {25, 25}, {25, 25}, {25, 26}, {26, 25}, {26, 26}, {26, 26}, {26, 27}, {27, 26},
+    {24, 32}, {27, 27}, {27, 28}, {28, 27}, {25, 33}, {28, 28}, {28, 29}, {29, 28}, {27, 32},
+    {29, 29}, {29, 30}, {30, 29}, {28, 33}, {30, 30}, {30, 31}, {31, 30}, {30, 32}, {31, 31},
+    {31, 32}, {32, 30}, {32, 31}, {33, 30}, {32, 32}, {32, 33}, {33, 32}, {34, 31}, {33, 33},
+    {33, 34}, {34, 33}, {36, 30}, {34, 34}, {34, 35}, {35, 34}, {37, 31}, {35, 35}, {35, 36},
+    {36, 35}, {36, 36}, {36, 36}, {36, 37}, {37, 36}, {37, 37}, {37, 37}, {37, 38}, {38, 37},
+    {38, 38}, {38, 38}, {38, 39}, {39, 38}, {39, 39}, {39, 39}, {39, 40}, {40, 39}, {40, 40},
+    {40, 40}, {40, 41}, {41, 40}, {41, 41}, {41, 41}, {41, 42}, {42, 41}, {42, 42}, {42, 42},
+    {42, 43}, {43, 42}, {40, 48}, {43, 43}, {43, 44}, {44, 43}, {41, 49}, {44, 44}, {44, 45},
+    {45, 44}, {43, 48}, {45, 45}, {45, 46}, {46, 45}, {44, 49}, {46, 46}, {46, 47}, {47, 46},
+    {46, 48}, {47, 47}, {47, 48}, {48, 46}, {48, 47}, {49, 46}, {48, 48}, {48, 49}, {49, 48},
+    {50, 47}, {49, 49}, {49, 50}, {50, 49}, {52, 46}, {50, 50}, {50, 51}, {51, 50}, {53, 47},
+    {51, 51}, {51, 52}, {52, 51}, {52, 52}, {52, 52}, {52, 53}, {53, 52}, {53, 53}, {53, 53},
+    {53, 54}, {54, 53}, {54, 54}, {54, 54}, {54, 55}, {55, 54}, {55, 55}, {55, 55}, {55, 56},
+    {56, 55}, {56, 56}, {56, 56}, {56, 57}, {57, 56}, {57, 57}, {57, 57}, {57, 58}, {58, 57},
+    {58, 58}, {58, 58}, {58, 59}, {59, 58}, {59, 59}, {59, 59}, {59, 60}, {60, 59}, {60, 60},
+    {60, 60}, {60, 61}, {61, 60}, {61, 61}, {61, 61}, {61, 62}, {62, 61}, {62, 62}, {62, 62},
+    {62, 63}, {63, 62}, {63, 63}, {63, 63},
+};
+
+static int stb__Mul8Bit(int a, int b) {
+    int t = a * b + 128;
+    return (t + (t >> 8)) >> 8;
+}
+
+static void stb__From16Bit(unsigned char* out, unsigned short v) {
+    int rv = (v & 0xf800) >> 11;
+    int gv = (v & 0x07e0) >> 5;
+    int bv = (v & 0x001f) >> 0;
+
+    // expand to 8 bits via bit replication
+    out[0] = static_cast<unsigned char>((rv * 33) >> 2);
+    out[1] = static_cast<unsigned char>((gv * 65) >> 4);
+    out[2] = static_cast<unsigned char>((bv * 33) >> 2);
+    out[3] = 0;
+}
+
+static unsigned short stb__As16Bit(int r, int g, int b) {
+    return (unsigned short)((stb__Mul8Bit(r, 31) << 11) + (stb__Mul8Bit(g, 63) << 5) +
+                            stb__Mul8Bit(b, 31));
+}
+
+// linear interpolation at 1/3 point between a and b, using desired rounding
+// type
+static int stb__Lerp13(int a, int b) {
+#ifdef STB_DXT_USE_ROUNDING_BIAS
+    // with rounding bias
+    return a + stb__Mul8Bit(b - a, 0x55);
+#else
+    // without rounding bias
+    // replace "/ 3" by "* 0xaaab) >> 17" if your compiler sucks or you really
+    // need every ounce of speed.
+    return (2 * a + b) / 3;
+#endif
+}
+
+// linear interpolation at 1/2 point between a and b
+static int stb__Lerp12(int a, int b) {
+    return (a + b) / 2;
+}
+
+// lerp RGB color
+static void stb__Lerp13RGB(unsigned char* out, unsigned char* p1, unsigned char* p2) {
+    out[0] = (unsigned char)stb__Lerp13(p1[0], p2[0]);
+    out[1] = (unsigned char)stb__Lerp13(p1[1], p2[1]);
+    out[2] = (unsigned char)stb__Lerp13(p1[2], p2[2]);
+}
+
+static void stb__Lerp12RGB(unsigned char* out, unsigned char* p1, unsigned char* p2) {
+    out[0] = (unsigned char)stb__Lerp12(p1[0], p2[0]);
+    out[1] = (unsigned char)stb__Lerp12(p1[1], p2[1]);
+    out[2] = (unsigned char)stb__Lerp12(p1[2], p2[2]);
+}
+
+/****************************************************************************/
+
+static void stb__Eval4Colors(unsigned char* color, unsigned short c0, unsigned short c1) {
+    stb__From16Bit(color + 0, c0);
+    stb__From16Bit(color + 4, c1);
+    stb__Lerp13RGB(color + 8, color + 0, color + 4);
+    stb__Lerp13RGB(color + 12, color + 4, color + 0);
+}
+
+static void stb__Eval3Colors(unsigned char* color, unsigned short c0, unsigned short c1) {
+    stb__From16Bit(color + 0, c0);
+    stb__From16Bit(color + 4, c1);
+    stb__Lerp12RGB(color + 8, color + 0, color + 4);
+}
+
+// The color matching function
+static unsigned int stb__MatchColorsBlock(unsigned char* block, unsigned char* color) {
+    unsigned int mask = 0;
+    int dirr = color[0 * 4 + 0] - color[1 * 4 + 0];
+    int dirg = color[0 * 4 + 1] - color[1 * 4 + 1];
+    int dirb = color[0 * 4 + 2] - color[1 * 4 + 2];
+    int dots[16];
+    int stops[4];
+    int i;
+    int c0Point, halfPoint, c3Point;
+
+    for (i = 0; i < 16; i++)
+        dots[i] = block[i * 4 + 0] * dirr + block[i * 4 + 1] * dirg + block[i * 4 + 2] * dirb;
+
+    for (i = 0; i < 4; i++)
+        stops[i] = color[i * 4 + 0] * dirr + color[i * 4 + 1] * dirg + color[i * 4 + 2] * dirb;
+
+    // think of the colors as arranged on a line; project point onto that line,
+    // then choose next color out of available ones. we compute the crossover
+    // points for "best color in top half"/"best in bottom half" and then the same
+    // inside that subinterval.
+    //
+    // relying on this 1d approximation isn't always optimal in terms of euclidean
+    // distance, but it's very close and a lot faster.
+    // http://cbloomrants.blogspot.com/2008/12/12-08-08-dxtc-summary.html
+
+    c0Point = (stops[1] + stops[3]);
+    halfPoint = (stops[3] + stops[2]);
+    c3Point = (stops[2] + stops[0]);
+
+    for (i = 15; i >= 0; i--) {
+        int dot = dots[i] * 2;
+        mask <<= 2;
+
+        if (dot < halfPoint)
+            mask |= (dot < c0Point) ? 1 : 3;
+        else
+            mask |= (dot < c3Point) ? 2 : 0;
+    }
+
+    return mask;
+}
+
+static unsigned int stb__MatchColorsAlphaBlock(unsigned char* block, unsigned char* color) {
+    unsigned int mask = 0;
+    int dirr = color[0 * 4 + 0] - color[1 * 4 + 0];
+    int dirg = color[0 * 4 + 1] - color[1 * 4 + 1];
+    int dirb = color[0 * 4 + 2] - color[1 * 4 + 2];
+    int dots[16];
+    int stops[3];
+    int i;
+    int c0Point, c2Point;
+
+    for (i = 0; i < 16; i++)
+        dots[i] = block[i * 4 + 0] * dirr + block[i * 4 + 1] * dirg + block[i * 4 + 2] * dirb;
+
+    for (i = 0; i < 3; i++)
+        stops[i] = color[i * 4 + 0] * dirr + color[i * 4 + 1] * dirg + color[i * 4 + 2] * dirb;
+
+    c0Point = (stops[1] + stops[2]);
+    c2Point = (stops[2] + stops[0]);
+
+    for (i = 15; i >= 0; i--) {
+        int dot = dots[i] * 2;
+        mask <<= 2;
+
+        if (block[i * 4 + 3] == 0)
+            mask |= 3;
+        else if (dot < c2Point)
+            mask |= (dot < c0Point) ? 0 : 2;
+        else
+            mask |= (dot < c0Point) ? 1 : 0;
+    }
+
+    return mask;
+}
+
+static void stb__ReorderColors(unsigned short* pmax16, unsigned short* pmin16) {
+    if (*pmin16 < *pmax16) {
+        unsigned short t = *pmin16;
+        *pmin16 = *pmax16;
+        *pmax16 = t;
+    }
+}
+
+static void stb__FinalizeColors(unsigned short* pmax16, unsigned short* pmin16,
+                                unsigned int* pmask) {
+    if (*pmax16 < *pmin16) {
+        unsigned short t = *pmin16;
+        *pmin16 = *pmax16;
+        *pmax16 = t;
+        *pmask ^= 0x55555555;
+    }
+}
+
+// The color optimization function. (Clever code, part 1)
+static void stb__OptimizeColorsBlock(unsigned char* block, unsigned short* pmax16,
+                                     unsigned short* pmin16) {
+    int mind, maxd;
+    unsigned char *minp, *maxp;
+    double magn;
+    int v_r, v_g, v_b;
+    static const int nIterPower = 4;
+    float covf[6], vfr, vfg, vfb;
+
+    // determine color distribution
+    int cov[6];
+    int mu[3], min[3], max[3];
+    int ch, i, iter;
+
+    for (ch = 0; ch < 3; ch++) {
+        const unsigned char* bp = ((const unsigned char*)block) + ch;
+        int muv, minv, maxv;
+
+        muv = minv = maxv = bp[0];
+        for (i = 4; i < 64; i += 4) {
+            muv += bp[i];
+            if (bp[i] < minv)
+                minv = bp[i];
+            else if (bp[i] > maxv)
+                maxv = bp[i];
+        }
+
+        mu[ch] = (muv + 8) >> 4;
+        min[ch] = minv;
+        max[ch] = maxv;
+    }
+
+    // determine covariance matrix
+    for (i = 0; i < 6; i++)
+        cov[i] = 0;
+
+    for (i = 0; i < 16; i++) {
+        int r = block[i * 4 + 0] - mu[0];
+        int g = block[i * 4 + 1] - mu[1];
+        int b = block[i * 4 + 2] - mu[2];
+
+        cov[0] += r * r;
+        cov[1] += r * g;
+        cov[2] += r * b;
+        cov[3] += g * g;
+        cov[4] += g * b;
+        cov[5] += b * b;
+    }
+
+    // convert covariance matrix to float, find principal axis via power iter
+    for (i = 0; i < 6; i++)
+        covf[i] = static_cast<float>(cov[i]) / 255.0f;
+
+    vfr = (float)(max[0] - min[0]);
+    vfg = (float)(max[1] - min[1]);
+    vfb = (float)(max[2] - min[2]);
+
+    for (iter = 0; iter < nIterPower; iter++) {
+        float r = vfr * covf[0] + vfg * covf[1] + vfb * covf[2];
+        float g = vfr * covf[1] + vfg * covf[3] + vfb * covf[4];
+        float b = vfr * covf[2] + vfg * covf[4] + vfb * covf[5];
+
+        vfr = r;
+        vfg = g;
+        vfb = b;
+    }
+
+    magn = STBD_FABS(vfr);
+    if (STBD_FABS(vfg) > magn)
+        magn = STBD_FABS(vfg);
+    if (STBD_FABS(vfb) > magn)
+        magn = STBD_FABS(vfb);
+
+    if (magn < 4.0f) { // too small, default to luminance
+        v_r = 299;     // JPEG YCbCr luma coefs, scaled by 1000.
+        v_g = 587;
+        v_b = 114;
+    } else {
+        magn = 512.0 / magn;
+        v_r = (int)(vfr * magn);
+        v_g = (int)(vfg * magn);
+        v_b = (int)(vfb * magn);
+    }
+
+    minp = maxp = block;
+    mind = maxd = block[0] * v_r + block[1] * v_g + block[2] * v_b;
+    // Pick colors at extreme points
+    for (i = 1; i < 16; i++) {
+        int dot = block[i * 4 + 0] * v_r + block[i * 4 + 1] * v_g + block[i * 4 + 2] * v_b;
+
+        if (dot < mind) {
+            mind = dot;
+            minp = block + i * 4;
+        }
+
+        if (dot > maxd) {
+            maxd = dot;
+            maxp = block + i * 4;
+        }
+    }
+
+    *pmax16 = stb__As16Bit(maxp[0], maxp[1], maxp[2]);
+    *pmin16 = stb__As16Bit(minp[0], minp[1], minp[2]);
+    stb__ReorderColors(pmax16, pmin16);
+}
+
+static void stb__OptimizeColorsAlphaBlock(unsigned char* block, unsigned short* pmax16,
+                                          unsigned short* pmin16) {
+    int mind, maxd;
+    unsigned char *minp, *maxp;
+    double magn;
+    int v_r, v_g, v_b;
+    static const int nIterPower = 4;
+    float covf[6], vfr, vfg, vfb;
+
+    // determine color distribution
+    int cov[6];
+    int mu[3], min[3], max[3];
+    int ch, i, iter;
+
+    for (ch = 0; ch < 3; ch++) {
+        const unsigned char* bp = ((const unsigned char*)block) + ch;
+        int muv = 0, minv = 256, maxv = -1;
+        int num = 0;
+
+        for (i = 0; i < 64; i += 4) {
+            if (bp[3 - ch] == 0) {
+                continue;
+            }
+
+            muv += bp[i];
+            if (bp[i] < minv)
+                minv = bp[i];
+            else if (bp[i] > maxv)
+                maxv = bp[i];
+
+            num++;
+        }
+
+        mu[ch] = num > 0 ? (muv + 8) / num : 0;
+        min[ch] = minv;
+        max[ch] = maxv;
+    }
+
+    // determine covariance matrix
+    for (i = 0; i < 6; i++)
+        cov[i] = 0;
+
+    for (i = 0; i < 16; i++) {
+        if (block[i * 4 + 3] == 0) {
+            continue;
+        }
+
+        int r = block[i * 4 + 0] - mu[0];
+        int g = block[i * 4 + 1] - mu[1];
+        int b = block[i * 4 + 2] - mu[2];
+
+        cov[0] += r * r;
+        cov[1] += r * g;
+        cov[2] += r * b;
+        cov[3] += g * g;
+        cov[4] += g * b;
+        cov[5] += b * b;
+    }
+
+    // convert covariance matrix to float, find principal axis via power iter
+    for (i = 0; i < 6; i++)
+        covf[i] = static_cast<float>(cov[i]) / 255.0f;
+
+    vfr = (float)(max[0] - min[0]);
+    vfg = (float)(max[1] - min[1]);
+    vfb = (float)(max[2] - min[2]);
+
+    for (iter = 0; iter < nIterPower; iter++) {
+        float r = vfr * covf[0] + vfg * covf[1] + vfb * covf[2];
+        float g = vfr * covf[1] + vfg * covf[3] + vfb * covf[4];
+        float b = vfr * covf[2] + vfg * covf[4] + vfb * covf[5];
+
+        vfr = r;
+        vfg = g;
+        vfb = b;
+    }
+
+    magn = STBD_FABS(vfr);
+    if (STBD_FABS(vfg) > magn)
+        magn = STBD_FABS(vfg);
+    if (STBD_FABS(vfb) > magn)
+        magn = STBD_FABS(vfb);
+
+    if (magn < 4.0f) { // too small, default to luminance
+        v_r = 299;     // JPEG YCbCr luma coefs, scaled by 1000.
+        v_g = 587;
+        v_b = 114;
+    } else {
+        magn = 512.0 / magn;
+        v_r = (int)(vfr * magn);
+        v_g = (int)(vfg * magn);
+        v_b = (int)(vfb * magn);
+    }
+
+    minp = maxp = NULL;
+    mind = 0x7fffffff;
+    maxd = -0x80000000;
+
+    // Pick colors at extreme points
+    for (i = 0; i < 16; i++) {
+        if (block[i * 4 + 3] == 0) {
+            continue;
+        }
+
+        int dot = block[i * 4 + 0] * v_r + block[i * 4 + 1] * v_g + block[i * 4 + 2] * v_b;
+
+        if (dot < mind) {
+            mind = dot;
+            minp = block + i * 4;
+        }
+
+        if (dot > maxd) {
+            maxd = dot;
+            maxp = block + i * 4;
+        }
+    }
+
+    if (!maxp) {
+        // all alpha, no color
+        *pmin16 = 0xffff;
+        *pmax16 = 0;
+    } else {
+        // endpoint colors found
+        *pmax16 = stb__As16Bit(maxp[0], maxp[1], maxp[2]);
+        *pmin16 = stb__As16Bit(minp[0], minp[1], minp[2]);
+
+        if (*pmax16 == *pmin16) {
+            // modify the endpoints to indicate presence of an alpha block
+            if (*pmax16 > 0) {
+                (*pmax16)--;
+            } else {
+                (*pmin16)++;
+            }
+        }
+
+        stb__ReorderColors(pmax16, pmin16);
+    }
+}
+
+static const float stb__midpoints5[32] = {
+    0.015686f, 0.047059f, 0.078431f, 0.111765f, 0.145098f, 0.176471f, 0.207843f, 0.241176f,
+    0.274510f, 0.305882f, 0.337255f, 0.370588f, 0.403922f, 0.435294f, 0.466667f, 0.5f,
+    0.533333f, 0.564706f, 0.596078f, 0.629412f, 0.662745f, 0.694118f, 0.725490f, 0.758824f,
+    0.792157f, 0.823529f, 0.854902f, 0.888235f, 0.921569f, 0.952941f, 0.984314f, 1.0f};
+
+static const float stb__midpoints6[64] = {
+    0.007843f, 0.023529f, 0.039216f, 0.054902f, 0.070588f, 0.086275f, 0.101961f, 0.117647f,
+    0.133333f, 0.149020f, 0.164706f, 0.180392f, 0.196078f, 0.211765f, 0.227451f, 0.245098f,
+    0.262745f, 0.278431f, 0.294118f, 0.309804f, 0.325490f, 0.341176f, 0.356863f, 0.372549f,
+    0.388235f, 0.403922f, 0.419608f, 0.435294f, 0.450980f, 0.466667f, 0.482353f, 0.500000f,
+    0.517647f, 0.533333f, 0.549020f, 0.564706f, 0.580392f, 0.596078f, 0.611765f, 0.627451f,
+    0.643137f, 0.658824f, 0.674510f, 0.690196f, 0.705882f, 0.721569f, 0.737255f, 0.754902f,
+    0.772549f, 0.788235f, 0.803922f, 0.819608f, 0.835294f, 0.850980f, 0.866667f, 0.882353f,
+    0.898039f, 0.913725f, 0.929412f, 0.945098f, 0.960784f, 0.976471f, 0.992157f, 1.0f};
+
+static unsigned short stb__Quantize5(float x) {
+    unsigned short q;
+    x = x < 0 ? 0 : x > 1 ? 1 : x; // saturate
+    q = (unsigned short)(x * 31);
+    q += (x > stb__midpoints5[q]);
+    return q;
+}
+
+static unsigned short stb__Quantize6(float x) {
+    unsigned short q;
+    x = x < 0 ? 0 : x > 1 ? 1 : x; // saturate
+    q = (unsigned short)(x * 63);
+    q += (x > stb__midpoints6[q]);
+    return q;
+}
+
+// The refinement function. (Clever code, part 2)
+// Tries to optimize colors to suit block contents better.
+// (By solving a least squares system via normal equations+Cramer's rule)
+static int stb__RefineBlock(unsigned char* block, unsigned short* pmax16, unsigned short* pmin16,
+                            unsigned int mask) {
+    static const int w1Tab[4] = {3, 0, 2, 1};
+    static const int prods[4] = {0x090000, 0x000900, 0x040102, 0x010402};
+    // ^some magic to save a lot of multiplies in the accumulating loop...
+    // (precomputed products of weights for least squares system, accumulated
+    // inside one 32-bit register)
+
+    float f;
+    unsigned short oldMin, oldMax, min16, max16;
+    int i, akku = 0, xx, xy, yy;
+    int At1_r, At1_g, At1_b;
+    int At2_r, At2_g, At2_b;
+    unsigned int cm = mask;
+
+    oldMin = *pmin16;
+    oldMax = *pmax16;
+
+    if ((mask ^ (mask << 2)) < 4) // all pixels have the same index?
+    {
+        // yes, linear system would be singular; solve using optimal
+        // single-color match on average color
+        int r = 8, g = 8, b = 8;
+        for (i = 0; i < 16; ++i) {
+            r += block[i * 4 + 0];
+            g += block[i * 4 + 1];
+            b += block[i * 4 + 2];
+        }
+
+        r >>= 4;
+        g >>= 4;
+        b >>= 4;
+
+        max16 = static_cast<unsigned short>((stb__OMatch5[r][0] << 11) | (stb__OMatch6[g][0] << 5) |
+                                            stb__OMatch5[b][0]);
+        min16 = static_cast<unsigned short>((stb__OMatch5[r][1] << 11) | (stb__OMatch6[g][1] << 5) |
+                                            stb__OMatch5[b][1]);
+    } else {
+        At1_r = At1_g = At1_b = 0;
+        At2_r = At2_g = At2_b = 0;
+        for (i = 0; i < 16; ++i, cm >>= 2) {
+            int step = cm & 3;
+            int w1 = w1Tab[step];
+            int r = block[i * 4 + 0];
+            int g = block[i * 4 + 1];
+            int b = block[i * 4 + 2];
+
+            akku += prods[step];
+            At1_r += w1 * r;
+            At1_g += w1 * g;
+            At1_b += w1 * b;
+            At2_r += r;
+            At2_g += g;
+            At2_b += b;
+        }
+
+        At2_r = 3 * At2_r - At1_r;
+        At2_g = 3 * At2_g - At1_g;
+        At2_b = 3 * At2_b - At1_b;
+
+        // extract solutions and decide solvability
+        xx = akku >> 16;
+        yy = (akku >> 8) & 0xff;
+        xy = (akku >> 0) & 0xff;
+
+        f = 3.0f / 255.0f / static_cast<float>(xx * yy - xy * xy);
+
+        max16 = static_cast<unsigned short>(
+            stb__Quantize5(static_cast<float>(At1_r * yy - At2_r * xy) * f) << 11);
+        max16 |= static_cast<unsigned short>(
+            stb__Quantize6(static_cast<float>(At1_g * yy - At2_g * xy) * f) << 5);
+        max16 |= static_cast<unsigned short>(
+            stb__Quantize5(static_cast<float>(At1_b * yy - At2_b * xy) * f) << 0);
+
+        min16 = static_cast<unsigned short>(
+            stb__Quantize5(static_cast<float>(At2_r * xx - At1_r * xy) * f) << 11);
+        min16 |= static_cast<unsigned short>(
+            stb__Quantize6(static_cast<float>(At2_g * xx - At1_g * xy) * f) << 5);
+        min16 |= static_cast<unsigned short>(
+            stb__Quantize5(static_cast<float>(At2_b * xx - At1_b * xy) * f) << 0);
+    }
+
+    *pmin16 = min16;
+    *pmax16 = max16;
+    stb__ReorderColors(pmax16, pmin16);
+
+    return oldMin != min16 || oldMax != max16;
+}
+
+// Color block compression
+static void stb__CompressColorBlock(unsigned char* dest, unsigned char* block, int alpha,
+                                    int mode) {
+    unsigned int mask;
+    int i;
+    int refinecount;
+    unsigned short max16, min16;
+    unsigned char color[4 * 4];
+
+    refinecount = (mode & STB_DXT_HIGHQUAL) ? 2 : 1;
+
+    // check if block is constant
+    for (i = 1; i < 16; i++)
+        if (((unsigned int*)block)[i] != ((unsigned int*)block)[0])
+            break;
+
+    if (i == 16 && block[3] == 0 && alpha) { // constant alpha
+        mask = 0xffffffff;
+        max16 = 0;
+        min16 = 0xffff;
+    } else if (i == 16) { // constant color
+        int r = block[0], g = block[1], b = block[2];
+        mask = 0xaaaaaaaa;
+        max16 = static_cast<unsigned short>((stb__OMatch5[r][0] << 11) | (stb__OMatch6[g][0] << 5) |
+                                            stb__OMatch5[b][0]);
+        min16 = static_cast<unsigned short>((stb__OMatch5[r][1] << 11) | (stb__OMatch6[g][1] << 5) |
+                                            stb__OMatch5[b][1]);
+    } else if (alpha) {
+        stb__OptimizeColorsAlphaBlock(block, &max16, &min16);
+        stb__Eval3Colors(color, max16, min16);
+        mask = stb__MatchColorsAlphaBlock(block, color);
+    } else {
+        // first step: PCA+map along principal axis
+        stb__OptimizeColorsBlock(block, &max16, &min16);
+        if (max16 != min16) {
+            stb__Eval4Colors(color, max16, min16);
+            mask = stb__MatchColorsBlock(block, color);
+        } else
+            mask = 0;
+
+        // third step: refine (multiple times if requested)
+        for (i = 0; i < refinecount; i++) {
+            unsigned int lastmask = mask;
+
+            if (stb__RefineBlock(block, &max16, &min16, mask)) {
+                if (max16 != min16) {
+                    stb__Eval4Colors(color, max16, min16);
+                    mask = stb__MatchColorsBlock(block, color);
+                } else {
+                    mask = 0;
+                    break;
+                }
+            }
+
+            if (mask == lastmask)
+                break;
+        }
+    }
+
+    // write the color block
+    if (!alpha)
+        stb__FinalizeColors(&max16, &min16, &mask);
+
+    dest[0] = (unsigned char)(max16);
+    dest[1] = (unsigned char)(max16 >> 8);
+    dest[2] = (unsigned char)(min16);
+    dest[3] = (unsigned char)(min16 >> 8);
+    dest[4] = (unsigned char)(mask);
+    dest[5] = (unsigned char)(mask >> 8);
+    dest[6] = (unsigned char)(mask >> 16);
+    dest[7] = (unsigned char)(mask >> 24);
+}
+
+// Alpha block compression (this is easy for a change)
+static void stb__CompressAlphaBlock(unsigned char* dest, unsigned char* src, int stride) {
+    int i, dist, bias, dist4, dist2, bits, mask;
+
+    // find min/max color
+    int mn, mx;
+    mn = mx = src[0];
+
+    for (i = 1; i < 16; i++) {
+        if (src[i * stride] < mn)
+            mn = src[i * stride];
+        else if (src[i * stride] > mx)
+            mx = src[i * stride];
+    }
+
+    // encode them
+    dest[0] = (unsigned char)mx;
+    dest[1] = (unsigned char)mn;
+    dest += 2;
+
+    // determine bias and emit color indices
+    // given the choice of mx/mn, these indices are optimal:
+    // http://fgiesen.wordpress.com/2009/12/15/dxt5-alpha-block-index-determination/
+    dist = mx - mn;
+    dist4 = dist * 4;
+    dist2 = dist * 2;
+    bias = (dist < 8) ? (dist - 1) : (dist / 2 + 2);
+    bias -= mn * 7;
+    bits = 0, mask = 0;
+
+    for (i = 0; i < 16; i++) {
+        int a = src[i * stride] * 7 + bias;
+        int ind, t;
+
+        // select index. this is a "linear scale" lerp factor between 0 (val=min)
+        // and 7 (val=max).
+        t = (a >= dist4) ? -1 : 0;
+        ind = t & 4;
+        a -= dist4 & t;
+        t = (a >= dist2) ? -1 : 0;
+        ind += t & 2;
+        a -= dist2 & t;
+        ind += (a >= dist);
+
+        // turn linear scale into DXT index (0/1 are extremal pts)
+        ind = -ind & 7;
+        ind ^= (2 > ind);
+
+        // write index
+        mask |= ind << bits;
+        if ((bits += 3) >= 8) {
+            *dest++ = (unsigned char)mask;
+            mask >>= 8;
+            bits -= 8;
+        }
+    }
+}
+
+void stb_compress_bc1_block(unsigned char* dest, const unsigned char* src, int alpha, int mode) {
+    stb__CompressColorBlock(dest, (unsigned char*)src, alpha, mode);
+}
+
+void stb_compress_bc3_block(unsigned char* dest, const unsigned char* src, int mode) {
+    unsigned char data[16][4];
+    int i;
+
+    stb__CompressAlphaBlock(dest, (unsigned char*)src + 3, 4);
+    dest += 8;
+    // make a new copy of the data in which alpha is opaque,
+    // because code uses a fast test for color constancy
+    memcpy(data, src, 4 * 16);
+    for (i = 0; i < 16; ++i)
+        data[i][3] = 255;
+    src = &data[0][0];
+
+    stb__CompressColorBlock(dest, (unsigned char*)src, 0, mode);
+}

externals/stb/stb_dxt.h

@@ -0,0 +1,36 @@
+// SPDX-FileCopyrightText: fabian "ryg" giesen
+// SPDX-License-Identifier: MIT
+
+// stb_dxt.h - v1.12 - DXT1/DXT5 compressor
+
+#ifndef STB_INCLUDE_STB_DXT_H
+#define STB_INCLUDE_STB_DXT_H
+
+#ifdef __cplusplus
+extern "C" {
+#endif
+
+#ifdef STB_DXT_STATIC
+#define STBDDEF static
+#else
+#define STBDDEF extern
+#endif
+
+// compression mode (bitflags)
+#define STB_DXT_NORMAL 0
+#define STB_DXT_DITHER 1 // use dithering. was always dubious, now deprecated. does nothing!
+#define STB_DXT_HIGHQUAL                                                                           \
+    2 // high quality mode, does two refinement steps instead of 1. ~30-40% slower.
+
+STBDDEF void stb_compress_bc1_block(unsigned char* dest,
+                                    const unsigned char* src_rgba_four_bytes_per_pixel, int alpha,
+                                    int mode);
+
+STBDDEF void stb_compress_bc3_block(unsigned char* dest, const unsigned char* src, int mode);
+
+#define STB_COMPRESS_DXT_BLOCK
+
+#ifdef __cplusplus
+}
+#endif
+#endif // STB_INCLUDE_STB_DXT_H

src/CMakeLists.txt

@@ -83,7 +83,7 @@ if (MSVC)
     )
 
     if (USE_CCACHE OR YUZU_USE_PRECOMPILED_HEADERS)
-    # when caching, we need to use /Z7 to downgrade debug info to use an older but more cachable format
+    # when caching, we need to use /Z7 to downgrade debug info to use an older but more cacheable format
     # Precompiled headers are deleted if not using /Z7. See https://github.com/nanoant/CMakePCHCompiler/issues/21
         add_compile_options(/Z7)
     else()
@@ -113,6 +113,9 @@ else()
 
         $<$<CXX_COMPILER_ID:Clang>:-Wno-braced-scalar-init>
         $<$<CXX_COMPILER_ID:Clang>:-Wno-unused-private-field>
+        $<$<CXX_COMPILER_ID:Clang>:-Werror=shadow-uncaptured-local>
+        $<$<CXX_COMPILER_ID:Clang>:-Werror=implicit-fallthrough>
+        $<$<CXX_COMPILER_ID:Clang>:-Werror=type-limits>
         $<$<CXX_COMPILER_ID:AppleClang>:-Wno-braced-scalar-init>
         $<$<CXX_COMPILER_ID:AppleClang>:-Wno-unused-private-field>
     )
@@ -126,6 +129,17 @@ else()
         add_compile_options("-stdlib=libc++")
     endif()
 
+    # GCC bugs
+    if (CMAKE_CXX_COMPILER_VERSION VERSION_GREATER_EQUAL "12" AND CMAKE_CXX_COMPILER_ID STREQUAL "GNU")
+        # These diagnostics would be great if they worked, but are just completely broken
+        # and produce bogus errors on external libraries like fmt.
+        add_compile_options(
+            -Wno-array-bounds
+            -Wno-stringop-overread
+            -Wno-stringop-overflow
+        )
+    endif()
+
     # Set file offset size to 64 bits.
     #
     # On modern Unixes, this is typically already the case. The lone exception is

src/audio_core/audio_in_manager.cpp

@@ -20,7 +20,7 @@ Manager::Manager(Core::System& system_) : system{system_} {
 Result Manager::AcquireSessionId(size_t& session_id) {
     if (num_free_sessions == 0) {
         LOG_ERROR(Service_Audio, "All 4 AudioIn sessions are in use, cannot create any more");
-        return Service::Audio::ERR_MAXIMUM_SESSIONS_REACHED;
+        return Service::Audio::ResultOutOfSessions;
     }
     session_id = session_ids[next_session_id];
     next_session_id = (next_session_id + 1) % MaxInSessions;

src/audio_core/audio_manager.cpp

@@ -19,7 +19,7 @@ void AudioManager::Shutdown() {
 
 Result AudioManager::SetOutManager(BufferEventFunc buffer_func) {
     if (!running) {
-        return Service::Audio::ERR_OPERATION_FAILED;
+        return Service::Audio::ResultOperationFailed;
     }
 
     std::scoped_lock l{lock};
@@ -35,7 +35,7 @@ Result AudioManager::SetOutManager(BufferEventFunc buffer_func) {
 
 Result AudioManager::SetInManager(BufferEventFunc buffer_func) {
     if (!running) {
-        return Service::Audio::ERR_OPERATION_FAILED;
+        return Service::Audio::ResultOperationFailed;
     }
 
     std::scoped_lock l{lock};

src/audio_core/audio_out_manager.cpp

@@ -19,7 +19,7 @@ Manager::Manager(Core::System& system_) : system{system_} {
 Result Manager::AcquireSessionId(size_t& session_id) {
     if (num_free_sessions == 0) {
         LOG_ERROR(Service_Audio, "All 12 Audio Out sessions are in use, cannot create any more");
-        return Service::Audio::ERR_MAXIMUM_SESSIONS_REACHED;
+        return Service::Audio::ResultOutOfSessions;
     }
     session_id = session_ids[next_session_id];
     next_session_id = (next_session_id + 1) % MaxOutSessions;

src/audio_core/audio_out_manager.h

@@ -58,7 +58,7 @@ public:
     /**
      * Get a list of audio out device names.
      *
-     * @oaram names     - Output container to write names to.
+     * @param names     - Output container to write names to.
      * @return Number of names written.
      */
     u32 GetAudioOutDeviceNames(

src/audio_core/audio_render_manager.cpp

@@ -28,7 +28,7 @@ SystemManager& Manager::GetSystemManager() {
 Result Manager::GetWorkBufferSize(const AudioRendererParameterInternal& params,
                                   u64& out_count) const {
     if (!CheckValidRevision(params.revision)) {
-        return Service::Audio::ERR_INVALID_REVISION;
+        return Service::Audio::ResultInvalidRevision;
     }
 
     out_count = System::GetWorkBufferSize(params);

src/audio_core/device/audio_buffer.h

@@ -16,7 +16,7 @@ struct AudioBuffer {
     s64 played_timestamp;
     /// Game memory address for these samples.
     VAddr samples;
-    /// Unqiue identifier for this buffer.
+    /// Unique identifier for this buffer.
     u64 tag;
     /// Size of the samples buffer.
     u64 size;

src/audio_core/device/device_session.cpp

@@ -93,7 +93,7 @@ void DeviceSession::AppendBuffers(std::span<const AudioBuffer> buffers) const {
             stream->AppendBuffer(new_buffer, samples);
         } else {
             std::vector<s16> samples(buffer.size / sizeof(s16));
-            system.Memory().ReadBlockUnsafe(buffer.samples, samples.data(), buffer.size);
+            system.ApplicationMemory().ReadBlockUnsafe(buffer.samples, samples.data(), buffer.size);
             stream->AppendBuffer(new_buffer, samples);
         }
     }
@@ -102,7 +102,7 @@ void DeviceSession::AppendBuffers(std::span<const AudioBuffer> buffers) const {
 void DeviceSession::ReleaseBuffer(const AudioBuffer& buffer) const {
     if (type == Sink::StreamType::In) {
         auto samples{stream->ReleaseBuffer(buffer.size / sizeof(s16))};
-        system.Memory().WriteBlockUnsafe(buffer.samples, samples.data(), buffer.size);
+        system.ApplicationMemory().WriteBlockUnsafe(buffer.samples, samples.data(), buffer.size);
     }
 }
 
@@ -121,8 +121,7 @@ u64 DeviceSession::GetPlayedSampleCount() const {
 }
 
 std::optional<std::chrono::nanoseconds> DeviceSession::ThreadFunc() {
-    // Add 5ms of samples at a 48K sample rate.
-    played_sample_count += 48'000 * INCREMENT_TIME / 1s;
+    played_sample_count = stream->GetExpectedPlayedSampleCount();
     if (type == Sink::StreamType::Out) {
         system.AudioCore().GetAudioManager().SetEvent(Event::Type::AudioOutManager, true);
     } else {

src/audio_core/in/audio_in.cpp

@@ -46,7 +46,7 @@ Result In::AppendBuffer(const AudioInBuffer& buffer, u64 tag) {
     if (system.AppendBuffer(buffer, tag)) {
         return ResultSuccess;
     }
-    return Service::Audio::ERR_BUFFER_COUNT_EXCEEDED;
+    return Service::Audio::ResultBufferCountReached;
 }
 
 void In::ReleaseAndRegisterBuffers() {

src/audio_core/in/audio_in_system.cpp

@@ -45,11 +45,11 @@ Result System::IsConfigValid(const std::string_view device_name,
                              const AudioInParameter& in_params) const {
     if ((device_name.size() > 0) &&
         (device_name != GetDefaultDeviceName() && device_name != GetDefaultUacDeviceName())) {
-        return Service::Audio::ERR_INVALID_DEVICE_NAME;
+        return Service::Audio::ResultNotFound;
     }
 
     if (in_params.sample_rate != TargetSampleRate && in_params.sample_rate > 0) {
-        return Service::Audio::ERR_INVALID_SAMPLE_RATE;
+        return Service::Audio::ResultInvalidSampleRate;
     }
 
     return ResultSuccess;
@@ -80,7 +80,7 @@ Result System::Initialize(std::string device_name, const AudioInParameter& in_pa
 
 Result System::Start() {
     if (state != State::Stopped) {
-        return Service::Audio::ERR_OPERATION_FAILED;
+        return Service::Audio::ResultOperationFailed;
     }
 
     session->Initialize(name, sample_format, channel_count, session_id, handle,

src/audio_core/out/audio_out.cpp

@@ -46,7 +46,7 @@ Result Out::AppendBuffer(const AudioOutBuffer& buffer, const u64 tag) {
     if (system.AppendBuffer(buffer, tag)) {
         return ResultSuccess;
     }
-    return Service::Audio::ERR_BUFFER_COUNT_EXCEEDED;
+    return Service::Audio::ResultBufferCountReached;
 }
 
 void Out::ReleaseAndRegisterBuffers() {

src/audio_core/out/audio_out_system.cpp

@@ -33,11 +33,11 @@ std::string_view System::GetDefaultOutputDeviceName() const {
 Result System::IsConfigValid(std::string_view device_name,
                              const AudioOutParameter& in_params) const {
     if ((device_name.size() > 0) && (device_name != GetDefaultOutputDeviceName())) {
-        return Service::Audio::ERR_INVALID_DEVICE_NAME;
+        return Service::Audio::ResultNotFound;
     }
 
     if (in_params.sample_rate != TargetSampleRate && in_params.sample_rate > 0) {
-        return Service::Audio::ERR_INVALID_SAMPLE_RATE;
+        return Service::Audio::ResultInvalidSampleRate;
     }
 
     if (in_params.channel_count == 0 || in_params.channel_count == 2 ||
@@ -45,7 +45,7 @@ Result System::IsConfigValid(std::string_view device_name,
         return ResultSuccess;
     }
 
-    return Service::Audio::ERR_INVALID_CHANNEL_COUNT;
+    return Service::Audio::ResultInvalidChannelCount;
 }
 
 Result System::Initialize(std::string device_name, const AudioOutParameter& in_params, u32 handle_,
@@ -80,7 +80,7 @@ size_t System::GetSessionId() const {
 
 Result System::Start() {
     if (state != State::Stopped) {
-        return Service::Audio::ERR_OPERATION_FAILED;
+        return Service::Audio::ResultOperationFailed;
     }
 
     session->Initialize(name, sample_format, channel_count, session_id, handle,

src/audio_core/renderer/adsp/adsp.cpp

@@ -13,7 +13,7 @@
 namespace AudioCore::AudioRenderer::ADSP {
 
 ADSP::ADSP(Core::System& system_, Sink::Sink& sink_)
-    : system{system_}, memory{system.Memory()}, sink{sink_} {}
+    : system{system_}, memory{system.ApplicationMemory()}, sink{sink_} {}
 
 ADSP::~ADSP() {
     ClearCommandBuffers();

src/audio_core/renderer/adsp/audio_renderer.cpp

@@ -132,10 +132,10 @@ void AudioRenderer::CreateSinkStreams() {
 }
 
 void AudioRenderer::ThreadFunc() {
-    constexpr char name[]{"AudioRenderer"};
+    static constexpr char name[]{"AudioRenderer"};
     MicroProfileOnThreadCreate(name);
     Common::SetCurrentThreadName(name);
-    Common::SetCurrentThreadPriority(Common::ThreadPriority::Critical);
+    Common::SetCurrentThreadPriority(Common::ThreadPriority::High);
     if (mailbox->ADSPWaitMessage() != RenderMessage::AudioRenderer_InitializeOK) {
         LOG_ERROR(Service_Audio,
                   "ADSP Audio Renderer -- Failed to receive initialize message from host!");
@@ -165,7 +165,7 @@ void AudioRenderer::ThreadFunc() {
                 // Check this buffer is valid, as it may not be used.
                 if (command_buffer.buffer != 0) {
                     // If there are no remaining commands (from the previous list),
-                    // this is a new command list, initalize it.
+                    // this is a new command list, initialize it.
                     if (command_buffer.remaining_command_count == 0) {
                         command_list_processor.Initialize(system, command_buffer.buffer,
                                                           command_buffer.size, streams[index]);
@@ -189,6 +189,8 @@ void AudioRenderer::ThreadFunc() {
                     max_time = std::min(command_buffer.time_limit, max_time);
                     command_list_processor.SetProcessTimeMax(max_time);
 
+                    streams[index]->WaitFreeSpace();
+
                     // Process the command list
                     {
                         MICROPROFILE_SCOPE(Audio_Renderer);

src/audio_core/renderer/adsp/audio_renderer.h

@@ -10,6 +10,7 @@
 #include "audio_core/renderer/adsp/command_buffer.h"
 #include "audio_core/renderer/adsp/command_list_processor.h"
 #include "common/common_types.h"
+#include "common/polyfill_thread.h"
 #include "common/reader_writer_queue.h"
 #include "common/thread.h"
 

src/audio_core/renderer/adsp/command_list_processor.cpp

@@ -17,7 +17,7 @@ namespace AudioCore::AudioRenderer::ADSP {
 void CommandListProcessor::Initialize(Core::System& system_, CpuAddr buffer, u64 size,
                                       Sink::SinkStream* stream_) {
     system = &system_;
-    memory = &system->Memory();
+    memory = &system->ApplicationMemory();
     stream = stream_;
     header = reinterpret_cast<CommandListHeader*>(buffer);
     commands = reinterpret_cast<u8*>(buffer + sizeof(CommandListHeader));

src/audio_core/renderer/audio_renderer.cpp

@@ -22,7 +22,7 @@ Result Renderer::Initialize(const AudioRendererParameterInternal& params,
         if (!manager.AddSystem(system)) {
             LOG_ERROR(Service_Audio,
                       "Both Audio Render sessions are in use, cannot create any more");
-            return Service::Audio::ERR_MAXIMUM_SESSIONS_REACHED;
+            return Service::Audio::ResultOutOfSessions;
         }
         system_registered = true;
     }

src/audio_core/renderer/behavior/behavior_info.h

@@ -155,7 +155,7 @@ public:
     /**
      * Check if a variadic command buffer is supported.
      * As of Rev 5 with the added optional performance metric logging, the command
-     * buffer can be a variable size, so take that into account for calcualting its size.
+     * buffer can be a variable size, so take that into account for calculating its size.
      *
      * @return True if supported, otherwise false.
      */

src/audio_core/renderer/behavior/info_updater.cpp

@@ -48,7 +48,7 @@ Result InfoUpdater::UpdateVoiceChannelResources(VoiceContext& voice_context) {
         LOG_ERROR(Service_Audio,
                   "Consumed an incorrect voice resource size, header size={}, consumed={}",
                   in_header->voice_resources_size, consumed_input_size);
-        return Service::Audio::ERR_INVALID_UPDATE_DATA;
+        return Service::Audio::ResultInvalidUpdateInfo;
     }
 
     input += consumed_input_size;
@@ -123,7 +123,7 @@ Result InfoUpdater::UpdateVoices(VoiceContext& voice_context,
     if (consumed_input_size != in_header->voices_size) {
         LOG_ERROR(Service_Audio, "Consumed an incorrect voices size, header size={}, consumed={}",
                   in_header->voices_size, consumed_input_size);
-        return Service::Audio::ERR_INVALID_UPDATE_DATA;
+        return Service::Audio::ResultInvalidUpdateInfo;
     }
 
     out_header->voices_size = consumed_output_size;
@@ -184,7 +184,7 @@ Result InfoUpdater::UpdateEffectsVersion1(EffectContext& effect_context, const b
     if (consumed_input_size != in_header->effects_size) {
         LOG_ERROR(Service_Audio, "Consumed an incorrect effects size, header size={}, consumed={}",
                   in_header->effects_size, consumed_input_size);
-        return Service::Audio::ERR_INVALID_UPDATE_DATA;
+        return Service::Audio::ResultInvalidUpdateInfo;
     }
 
     out_header->effects_size = consumed_output_size;
@@ -239,7 +239,7 @@ Result InfoUpdater::UpdateEffectsVersion2(EffectContext& effect_context, const b
     if (consumed_input_size != in_header->effects_size) {
         LOG_ERROR(Service_Audio, "Consumed an incorrect effects size, header size={}, consumed={}",
                   in_header->effects_size, consumed_input_size);
-        return Service::Audio::ERR_INVALID_UPDATE_DATA;
+        return Service::Audio::ResultInvalidUpdateInfo;
     }
 
     out_header->effects_size = consumed_output_size;
@@ -267,7 +267,7 @@ Result InfoUpdater::UpdateMixes(MixContext& mix_context, const u32 mix_buffer_co
     }
 
     if (mix_buffer_count == 0) {
-        return Service::Audio::ERR_INVALID_UPDATE_DATA;
+        return Service::Audio::ResultInvalidUpdateInfo;
     }
 
     std::span<const MixInfo::InParameter> in_params{
@@ -281,13 +281,13 @@ Result InfoUpdater::UpdateMixes(MixContext& mix_context, const u32 mix_buffer_co
             total_buffer_count += params.buffer_count;
             if (params.dest_mix_id > static_cast<s32>(mix_context.GetCount()) &&
                 params.dest_mix_id != UnusedMixId && params.mix_id != FinalMixId) {
-                return Service::Audio::ERR_INVALID_UPDATE_DATA;
+                return Service::Audio::ResultInvalidUpdateInfo;
             }
         }
     }
 
     if (total_buffer_count > mix_buffer_count) {
-        return Service::Audio::ERR_INVALID_UPDATE_DATA;
+        return Service::Audio::ResultInvalidUpdateInfo;
     }
 
     bool mix_dirty{false};
@@ -317,7 +317,7 @@ Result InfoUpdater::UpdateMixes(MixContext& mix_context, const u32 mix_buffer_co
     if (mix_dirty) {
         if (behaviour.IsSplitterSupported() && splitter_context.UsingSplitter()) {
             if (!mix_context.TSortInfo(splitter_context)) {
-                return Service::Audio::ERR_INVALID_UPDATE_DATA;
+                return Service::Audio::ResultInvalidUpdateInfo;
             }
         } else {
             mix_context.SortInfo();
@@ -327,7 +327,7 @@ Result InfoUpdater::UpdateMixes(MixContext& mix_context, const u32 mix_buffer_co
     if (consumed_input_size != in_header->mix_size) {
         LOG_ERROR(Service_Audio, "Consumed an incorrect mixes size, header size={}, consumed={}",
                   in_header->mix_size, consumed_input_size);
-        return Service::Audio::ERR_INVALID_UPDATE_DATA;
+        return Service::Audio::ResultInvalidUpdateInfo;
     }
 
     input += mix_count * sizeof(MixInfo::InParameter);
@@ -384,7 +384,7 @@ Result InfoUpdater::UpdateSinks(SinkContext& sink_context, std::span<MemoryPoolI
     if (consumed_input_size != in_header->sinks_size) {
         LOG_ERROR(Service_Audio, "Consumed an incorrect sinks size, header size={}, consumed={}",
                   in_header->sinks_size, consumed_input_size);
-        return Service::Audio::ERR_INVALID_UPDATE_DATA;
+        return Service::Audio::ResultInvalidUpdateInfo;
     }
 
     input += consumed_input_size;
@@ -411,7 +411,7 @@ Result InfoUpdater::UpdateMemoryPools(std::span<MemoryPoolInfo> memory_pools,
             state != MemoryPoolInfo::ResultState::MapFailed &&
             state != MemoryPoolInfo::ResultState::InUse) {
             LOG_WARNING(Service_Audio, "Invalid ResultState from updating memory pools");
-            return Service::Audio::ERR_INVALID_UPDATE_DATA;
+            return Service::Audio::ResultInvalidUpdateInfo;
         }
     }
 
@@ -423,7 +423,7 @@ Result InfoUpdater::UpdateMemoryPools(std::span<MemoryPoolInfo> memory_pools,
         LOG_ERROR(Service_Audio,
                   "Consumed an incorrect memory pool size, header size={}, consumed={}",
                   in_header->memory_pool_size, consumed_input_size);
-        return Service::Audio::ERR_INVALID_UPDATE_DATA;
+        return Service::Audio::ResultInvalidUpdateInfo;
     }
 
     input += consumed_input_size;
@@ -453,7 +453,7 @@ Result InfoUpdater::UpdatePerformanceBuffer(std::span<u8> performance_output,
         LOG_ERROR(Service_Audio,
                   "Consumed an incorrect performance size, header size={}, consumed={}",
                   in_header->performance_buffer_size, consumed_input_size);
-        return Service::Audio::ERR_INVALID_UPDATE_DATA;
+        return Service::Audio::ResultInvalidUpdateInfo;
     }
 
     input += consumed_input_size;
@@ -467,18 +467,18 @@ Result InfoUpdater::UpdateBehaviorInfo(BehaviorInfo& behaviour_) {
     const auto in_params{reinterpret_cast<const BehaviorInfo::InParameter*>(input)};
 
     if (!CheckValidRevision(in_params->revision)) {
-        return Service::Audio::ERR_INVALID_UPDATE_DATA;
+        return Service::Audio::ResultInvalidUpdateInfo;
     }
 
     if (in_params->revision != behaviour_.GetUserRevision()) {
-        return Service::Audio::ERR_INVALID_UPDATE_DATA;
+        return Service::Audio::ResultInvalidUpdateInfo;
     }
 
     behaviour_.ClearError();
     behaviour_.UpdateFlags(in_params->flags);
 
     if (in_header->behaviour_size != sizeof(BehaviorInfo::InParameter)) {
-        return Service::Audio::ERR_INVALID_UPDATE_DATA;
+        return Service::Audio::ResultInvalidUpdateInfo;
     }
 
     input += sizeof(BehaviorInfo::InParameter);
@@ -500,7 +500,7 @@ Result InfoUpdater::UpdateErrorInfo(const BehaviorInfo& behaviour_) {
 Result InfoUpdater::UpdateSplitterInfo(SplitterContext& splitter_context) {
     u32 consumed_size{0};
     if (!splitter_context.Update(input, consumed_size)) {
-        return Service::Audio::ERR_INVALID_UPDATE_DATA;
+        return Service::Audio::ResultInvalidUpdateInfo;
     }
 
     input += consumed_size;
@@ -529,9 +529,9 @@ Result InfoUpdater::UpdateRendererInfo(const u64 elapsed_frames) {
 
 Result InfoUpdater::CheckConsumedSize() {
     if (CpuAddr(input) - CpuAddr(input_origin.data()) != expected_input_size) {
-        return Service::Audio::ERR_INVALID_UPDATE_DATA;
+        return Service::Audio::ResultInvalidUpdateInfo;
     } else if (CpuAddr(output) - CpuAddr(output_origin.data()) != expected_output_size) {
-        return Service::Audio::ERR_INVALID_UPDATE_DATA;
+        return Service::Audio::ResultInvalidUpdateInfo;
     }
     return ResultSuccess;
 }

src/audio_core/renderer/command/command_buffer.cpp

@@ -251,8 +251,8 @@ void CommandBuffer::GenerateBiquadFilterCommand(const s32 node_id, EffectInfoBas
 
     const auto& parameter{
         *reinterpret_cast<BiquadFilterInfo::ParameterVersion1*>(effect_info.GetParameter())};
-    const auto state{
-        reinterpret_cast<VoiceState::BiquadFilterState*>(effect_info.GetStateBuffer())};
+    const auto state{reinterpret_cast<VoiceState::BiquadFilterState*>(
+        effect_info.GetStateBuffer() + channel * sizeof(VoiceState::BiquadFilterState))};
 
     cmd.input = buffer_offset + parameter.inputs[channel];
     cmd.output = buffer_offset + parameter.outputs[channel];

src/audio_core/renderer/command/command_generator.cpp

@@ -46,7 +46,7 @@ void CommandGenerator::GenerateDataSourceCommand(VoiceInfo& voice_info,
             while (destination != nullptr) {
                 if (destination->IsConfigured()) {
                     auto mix_id{destination->GetMixId()};
-                    if (mix_id < mix_context.GetCount()) {
+                    if (mix_id < mix_context.GetCount() && mix_id != UnusedSplitterId) {
                         auto mix_info{mix_context.GetInfo(mix_id)};
                         command_buffer.GenerateDepopPrepareCommand(
                             voice_info.node_id, voice_state, render_context.depop_buffer,

src/audio_core/renderer/command/effect/aux_.cpp

@@ -4,6 +4,7 @@
 #include "audio_core/renderer/adsp/command_list_processor.h"
 #include "audio_core/renderer/command/effect/aux_.h"
 #include "audio_core/renderer/effect/aux_.h"
+#include "core/core.h"
 #include "core/memory.h"
 
 namespace AudioCore::AudioRenderer {
@@ -19,10 +20,24 @@ static void ResetAuxBufferDsp(Core::Memory::Memory& memory, const CpuAddr aux_in
         return;
     }
 
-    auto info{reinterpret_cast<AuxInfo::AuxInfoDsp*>(memory.GetPointer(aux_info))};
-    info->read_offset = 0;
-    info->write_offset = 0;
-    info->total_sample_count = 0;
+    AuxInfo::AuxInfoDsp info{};
+    auto info_ptr{&info};
+    bool host_safe{(aux_info & Core::Memory::YUZU_PAGEMASK) <=
+                   (Core::Memory::YUZU_PAGESIZE - sizeof(AuxInfo::AuxInfoDsp))};
+
+    if (host_safe) [[likely]] {
+        info_ptr = memory.GetPointer<AuxInfo::AuxInfoDsp>(aux_info);
+    } else {
+        memory.ReadBlockUnsafe(aux_info, info_ptr, sizeof(AuxInfo::AuxInfoDsp));
+    }
+
+    info_ptr->read_offset = 0;
+    info_ptr->write_offset = 0;
+    info_ptr->total_sample_count = 0;
+
+    if (!host_safe) [[unlikely]] {
+        memory.WriteBlockUnsafe(aux_info, info_ptr, sizeof(AuxInfo::AuxInfoDsp));
+    }
 }
 
 /**
@@ -40,11 +55,10 @@ static void ResetAuxBufferDsp(Core::Memory::Memory& memory, const CpuAddr aux_in
  * @param update_count - If non-zero, send_info_ will be updated.
  * @return Number of samples written.
  */
-static u32 WriteAuxBufferDsp(Core::Memory::Memory& memory, const CpuAddr send_info_,
-                             [[maybe_unused]] u32 sample_count, const CpuAddr send_buffer,
-                             const u32 count_max, std::span<const s32> input,
-                             const u32 write_count_, const u32 write_offset,
-                             const u32 update_count) {
+static u32 WriteAuxBufferDsp(Core::Memory::Memory& memory, CpuAddr send_info_,
+                             [[maybe_unused]] u32 sample_count, CpuAddr send_buffer, u32 count_max,
+                             std::span<const s32> input, u32 write_count_, u32 write_offset,
+                             u32 update_count) {
     if (write_count_ > count_max) {
         LOG_ERROR(Service_Audio,
                   "write_count must be smaller than count_max! write_count {}, count_max {}",
@@ -52,6 +66,11 @@ static u32 WriteAuxBufferDsp(Core::Memory::Memory& memory, const CpuAddr send_in
         return 0;
     }
 
+    if (send_info_ == 0) {
+        LOG_ERROR(Service_Audio, "send_info_ is 0!");
+        return 0;
+    }
+
     if (input.empty()) {
         LOG_ERROR(Service_Audio, "input buffer is empty!");
         return 0;
@@ -67,33 +86,47 @@ static u32 WriteAuxBufferDsp(Core::Memory::Memory& memory, const CpuAddr send_in
     }
 
     AuxInfo::AuxInfoDsp send_info{};
-    memory.ReadBlockUnsafe(send_info_, &send_info, sizeof(AuxInfo::AuxInfoDsp));
+    auto send_ptr = &send_info;
+    bool host_safe = (send_info_ & Core::Memory::YUZU_PAGEMASK) <=
+                     (Core::Memory::YUZU_PAGESIZE - sizeof(AuxInfo::AuxInfoDsp));
 
-    u32 target_write_offset{send_info.write_offset + write_offset};
-    if (target_write_offset > count_max || write_count_ == 0) {
+    if (host_safe) [[likely]] {
+        send_ptr = memory.GetPointer<AuxInfo::AuxInfoDsp>(send_info_);
+    } else {
+        memory.ReadBlockUnsafe(send_info_, send_ptr, sizeof(AuxInfo::AuxInfoDsp));
+    }
+
+    u32 target_write_offset{send_ptr->write_offset + write_offset};
+    if (target_write_offset > count_max) {
         return 0;
     }
 
     u32 write_count{write_count_};
-    u32 write_pos{0};
+    u32 read_pos{0};
     while (write_count > 0) {
         u32 to_write{std::min(count_max - target_write_offset, write_count)};
-
-        if (to_write > 0) {
+        const auto write_addr = send_buffer + target_write_offset * sizeof(s32);
+        bool write_safe{(write_addr & Core::Memory::YUZU_PAGEMASK) <=
+                        (Core::Memory::YUZU_PAGESIZE - (write_addr + to_write * sizeof(s32)))};
+        if (write_safe) [[likely]] {
+            auto ptr = memory.GetPointer(write_addr);
+            std::memcpy(ptr, &input[read_pos], to_write * sizeof(s32));
+        } else {
             memory.WriteBlockUnsafe(send_buffer + target_write_offset * sizeof(s32),
-                                    &input[write_pos], to_write * sizeof(s32));
+                                    &input[read_pos], to_write * sizeof(s32));
         }
-
         target_write_offset = (target_write_offset + to_write) % count_max;
         write_count -= to_write;
-        write_pos += to_write;
+        read_pos += to_write;
     }
 
     if (update_count) {
-        send_info.write_offset = (send_info.write_offset + update_count) % count_max;
+        send_ptr->write_offset = (send_ptr->write_offset + update_count) % count_max;
     }
 
-    memory.WriteBlockUnsafe(send_info_, &send_info, sizeof(AuxInfo::AuxInfoDsp));
+    if (!host_safe) [[unlikely]] {
+        memory.WriteBlockUnsafe(send_info_, send_ptr, sizeof(AuxInfo::AuxInfoDsp));
+    }
 
     return write_count_;
 }
@@ -102,7 +135,7 @@ static u32 WriteAuxBufferDsp(Core::Memory::Memory& memory, const CpuAddr send_in
  * Read the given memory at return_buffer into the output mix buffer, and update return_info_ if
  * update_count is set, to notify the game that an update happened.
  *
- * @param memory        - Core memory for writing.
+ * @param memory        - Core memory for reading.
  * @param return_info_  - Meta information for where to read the mix buffer.
  * @param return_buffer - Memory address to read the samples from.
  * @param count_max     - Maximum number of samples in the receiving buffer.
@@ -112,16 +145,21 @@ static u32 WriteAuxBufferDsp(Core::Memory::Memory& memory, const CpuAddr send_in
  * @param update_count  - If non-zero, send_info_ will be updated.
  * @return Number of samples read.
  */
-static u32 ReadAuxBufferDsp(Core::Memory::Memory& memory, const CpuAddr return_info_,
-                            const CpuAddr return_buffer, const u32 count_max, std::span<s32> output,
-                            const u32 count_, const u32 read_offset, const u32 update_count) {
+static u32 ReadAuxBufferDsp(Core::Memory::Memory& memory, CpuAddr return_info_,
+                            CpuAddr return_buffer, u32 count_max, std::span<s32> output,
+                            u32 read_count_, u32 read_offset, u32 update_count) {
     if (count_max == 0) {
         return 0;
     }
 
-    if (count_ > count_max) {
+    if (read_count_ > count_max) {
         LOG_ERROR(Service_Audio, "count must be smaller than count_max! count {}, count_max {}",
-                  count_, count_max);
+                  read_count_, count_max);
+        return 0;
+    }
+
+    if (return_info_ == 0) {
+        LOG_ERROR(Service_Audio, "return_info_ is 0!");
         return 0;
     }
 
@@ -136,35 +174,49 @@ static u32 ReadAuxBufferDsp(Core::Memory::Memory& memory, const CpuAddr return_i
     }
 
     AuxInfo::AuxInfoDsp return_info{};
-    memory.ReadBlockUnsafe(return_info_, &return_info, sizeof(AuxInfo::AuxInfoDsp));
+    auto return_ptr = &return_info;
+    bool host_safe = (return_info_ & Core::Memory::YUZU_PAGEMASK) <=
+                     (Core::Memory::YUZU_PAGESIZE - sizeof(AuxInfo::AuxInfoDsp));
 
-    u32 target_read_offset{return_info.read_offset + read_offset};
+    if (host_safe) [[likely]] {
+        return_ptr = memory.GetPointer<AuxInfo::AuxInfoDsp>(return_info_);
+    } else {
+        memory.ReadBlockUnsafe(return_info_, return_ptr, sizeof(AuxInfo::AuxInfoDsp));
+    }
+
+    u32 target_read_offset{return_ptr->read_offset + read_offset};
     if (target_read_offset > count_max) {
         return 0;
     }
 
-    u32 read_count{count_};
-    u32 read_pos{0};
+    u32 read_count{read_count_};
+    u32 write_pos{0};
     while (read_count > 0) {
         u32 to_read{std::min(count_max - target_read_offset, read_count)};
-
-        if (to_read > 0) {
+        const auto read_addr = return_buffer + target_read_offset * sizeof(s32);
+        bool read_safe{(read_addr & Core::Memory::YUZU_PAGEMASK) <=
+                       (Core::Memory::YUZU_PAGESIZE - (read_addr + to_read * sizeof(s32)))};
+        if (read_safe) [[likely]] {
+            auto ptr = memory.GetPointer(read_addr);
+            std::memcpy(&output[write_pos], ptr, to_read * sizeof(s32));
+        } else {
             memory.ReadBlockUnsafe(return_buffer + target_read_offset * sizeof(s32),
-                                   &output[read_pos], to_read * sizeof(s32));
+                                   &output[write_pos], to_read * sizeof(s32));
         }
-
         target_read_offset = (target_read_offset + to_read) % count_max;
         read_count -= to_read;
-        read_pos += to_read;
+        write_pos += to_read;
     }
 
     if (update_count) {
-        return_info.read_offset = (return_info.read_offset + update_count) % count_max;
+        return_ptr->read_offset = (return_ptr->read_offset + update_count) % count_max;
     }
 
-    memory.WriteBlockUnsafe(return_info_, &return_info, sizeof(AuxInfo::AuxInfoDsp));
+    if (!host_safe) [[unlikely]] {
+        memory.WriteBlockUnsafe(return_info_, return_ptr, sizeof(AuxInfo::AuxInfoDsp));
+    }
 
-    return count_;
+    return read_count_;
 }
 
 void AuxCommand::Dump([[maybe_unused]] const ADSP::CommandListProcessor& processor,
@@ -189,7 +241,7 @@ void AuxCommand::Process(const ADSP::CommandListProcessor& processor) {
                                    update_count)};
 
         if (read != processor.sample_count) {
-            std::memset(&output_buffer[read], 0, processor.sample_count - read);
+            std::memset(&output_buffer[read], 0, (processor.sample_count - read) * sizeof(s32));
         }
     } else {
         ResetAuxBufferDsp(*processor.memory, send_buffer_info);

src/audio_core/renderer/command/effect/biquad_filter.cpp

@@ -4,6 +4,7 @@
 #include "audio_core/renderer/adsp/command_list_processor.h"
 #include "audio_core/renderer/command/effect/biquad_filter.h"
 #include "audio_core/renderer/voice/voice_state.h"
+#include "common/bit_cast.h"
 
 namespace AudioCore::AudioRenderer {
 /**
@@ -19,21 +20,21 @@ namespace AudioCore::AudioRenderer {
 void ApplyBiquadFilterFloat(std::span<s32> output, std::span<const s32> input,
                             std::array<s16, 3>& b_, std::array<s16, 2>& a_,
                             VoiceState::BiquadFilterState& state, const u32 sample_count) {
-    constexpr s64 min{std::numeric_limits<s32>::min()};
-    constexpr s64 max{std::numeric_limits<s32>::max()};
+    constexpr f64 min{std::numeric_limits<s32>::min()};
+    constexpr f64 max{std::numeric_limits<s32>::max()};
     std::array<f64, 3> b{Common::FixedPoint<50, 14>::from_base(b_[0]).to_double(),
                          Common::FixedPoint<50, 14>::from_base(b_[1]).to_double(),
                          Common::FixedPoint<50, 14>::from_base(b_[2]).to_double()};
     std::array<f64, 2> a{Common::FixedPoint<50, 14>::from_base(a_[0]).to_double(),
                          Common::FixedPoint<50, 14>::from_base(a_[1]).to_double()};
-    std::array<f64, 4> s{state.s0.to_double(), state.s1.to_double(), state.s2.to_double(),
-                         state.s3.to_double()};
+    std::array<f64, 4> s{Common::BitCast<f64>(state.s0), Common::BitCast<f64>(state.s1),
+                         Common::BitCast<f64>(state.s2), Common::BitCast<f64>(state.s3)};
 
     for (u32 i = 0; i < sample_count; i++) {
         f64 in_sample{static_cast<f64>(input[i])};
         auto sample{in_sample * b[0] + s[0] * b[1] + s[1] * b[2] + s[2] * a[0] + s[3] * a[1]};
 
-        output[i] = static_cast<s32>(std::clamp(static_cast<s64>(sample), min, max));
+        output[i] = static_cast<s32>(std::clamp(sample, min, max));
 
         s[1] = s[0];
         s[0] = in_sample;
@@ -41,10 +42,10 @@ void ApplyBiquadFilterFloat(std::span<s32> output, std::span<const s32> input,
         s[2] = sample;
     }
 
-    state.s0 = s[0];
-    state.s1 = s[1];
-    state.s2 = s[2];
-    state.s3 = s[3];
+    state.s0 = Common::BitCast<s64>(s[0]);
+    state.s1 = Common::BitCast<s64>(s[1]);
+    state.s2 = Common::BitCast<s64>(s[2]);
+    state.s3 = Common::BitCast<s64>(s[3]);
 }
 
 /**
@@ -58,29 +59,20 @@ void ApplyBiquadFilterFloat(std::span<s32> output, std::span<const s32> input,
  * @param sample_count - Number of samples to process.
  */
 static void ApplyBiquadFilterInt(std::span<s32> output, std::span<const s32> input,
-                                 std::array<s16, 3>& b_, std::array<s16, 2>& a_,
+                                 std::array<s16, 3>& b, std::array<s16, 2>& a,
                                  VoiceState::BiquadFilterState& state, const u32 sample_count) {
     constexpr s64 min{std::numeric_limits<s32>::min()};
     constexpr s64 max{std::numeric_limits<s32>::max()};
-    std::array<Common::FixedPoint<50, 14>, 3> b{
-        Common::FixedPoint<50, 14>::from_base(b_[0]),
-        Common::FixedPoint<50, 14>::from_base(b_[1]),
-        Common::FixedPoint<50, 14>::from_base(b_[2]),
-    };
-    std::array<Common::FixedPoint<50, 14>, 3> a{
-        Common::FixedPoint<50, 14>::from_base(a_[0]),
-        Common::FixedPoint<50, 14>::from_base(a_[1]),
-    };
 
     for (u32 i = 0; i < sample_count; i++) {
-        s64 in_sample{input[i]};
-        auto sample{in_sample * b[0] + state.s0};
-        const auto out_sample{std::clamp(sample.to_long(), min, max)};
+        const s64 in_sample{input[i]};
+        const s64 sample{in_sample * b[0] + state.s0};
+        const s64 out_sample{std::clamp<s64>((sample + (1 << 13)) >> 14, min, max)};
 
         output[i] = static_cast<s32>(out_sample);
 
         state.s0 = state.s1 + b[1] * in_sample + a[0] * out_sample;
-        state.s1 = 0 + b[2] * in_sample + a[1] * out_sample;
+        state.s1 = b[2] * in_sample + a[1] * out_sample;
     }
 }
 

src/audio_core/renderer/command/effect/i3dl2_reverb.cpp

@@ -244,16 +244,16 @@ template <size_t NumChannels>
 static void ApplyI3dl2ReverbEffect(I3dl2ReverbInfo::State& state,
                                    std::span<std::span<const s32>> inputs,
                                    std::span<std::span<s32>> outputs, const u32 sample_count) {
-    constexpr std::array<u8, I3dl2ReverbInfo::MaxDelayTaps> OutTapIndexes1Ch{
+    static constexpr std::array<u8, I3dl2ReverbInfo::MaxDelayTaps> OutTapIndexes1Ch{
         0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
     };
-    constexpr std::array<u8, I3dl2ReverbInfo::MaxDelayTaps> OutTapIndexes2Ch{
+    static constexpr std::array<u8, I3dl2ReverbInfo::MaxDelayTaps> OutTapIndexes2Ch{
         0, 0, 0, 1, 1, 1, 1, 0, 0, 0, 1, 1, 1, 0, 0, 0, 0, 1, 1, 1,
     };
-    constexpr std::array<u8, I3dl2ReverbInfo::MaxDelayTaps> OutTapIndexes4Ch{
+    static constexpr std::array<u8, I3dl2ReverbInfo::MaxDelayTaps> OutTapIndexes4Ch{
         0, 0, 0, 1, 1, 1, 1, 2, 2, 2, 1, 1, 1, 0, 0, 0, 0, 3, 3, 3,
     };
-    constexpr std::array<u8, I3dl2ReverbInfo::MaxDelayTaps> OutTapIndexes6Ch{
+    static constexpr std::array<u8, I3dl2ReverbInfo::MaxDelayTaps> OutTapIndexes6Ch{
         2, 0, 0, 1, 1, 1, 1, 4, 4, 4, 1, 1, 1, 0, 0, 0, 0, 5, 5, 5,
     };
 

src/audio_core/renderer/command/effect/reverb.cpp

@@ -252,16 +252,16 @@ template <size_t NumChannels>
 static void ApplyReverbEffect(const ReverbInfo::ParameterVersion2& params, ReverbInfo::State& state,
                               std::vector<std::span<const s32>>& inputs,
                               std::vector<std::span<s32>>& outputs, const u32 sample_count) {
-    constexpr std::array<u8, ReverbInfo::MaxDelayTaps> OutTapIndexes1Ch{
+    static constexpr std::array<u8, ReverbInfo::MaxDelayTaps> OutTapIndexes1Ch{
         0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
     };
-    constexpr std::array<u8, ReverbInfo::MaxDelayTaps> OutTapIndexes2Ch{
+    static constexpr std::array<u8, ReverbInfo::MaxDelayTaps> OutTapIndexes2Ch{
         0, 0, 1, 1, 0, 1, 0, 0, 1, 1,
     };
-    constexpr std::array<u8, ReverbInfo::MaxDelayTaps> OutTapIndexes4Ch{
+    static constexpr std::array<u8, ReverbInfo::MaxDelayTaps> OutTapIndexes4Ch{
         0, 0, 1, 1, 0, 1, 2, 2, 3, 3,
     };
-    constexpr std::array<u8, ReverbInfo::MaxDelayTaps> OutTapIndexes6Ch{
+    static constexpr std::array<u8, ReverbInfo::MaxDelayTaps> OutTapIndexes6Ch{
         0, 0, 1, 1, 2, 2, 4, 4, 5, 5,
     };
 
@@ -308,7 +308,8 @@ static void ApplyReverbEffect(const ReverbInfo::ParameterVersion2& params, Rever
         }
 
         Common::FixedPoint<50, 14> pre_delay_sample{
-            state.pre_delay_line.Read() * Common::FixedPoint<50, 14>::from_base(params.late_gain)};
+            state.pre_delay_line.TapOut(state.pre_delay_time) *
+            Common::FixedPoint<50, 14>::from_base(params.late_gain)};
 
         std::array<Common::FixedPoint<50, 14>, ReverbInfo::MaxDelayLines> mix_matrix{
             state.prev_feedback_output[2] + state.prev_feedback_output[1] + pre_delay_sample,

src/audio_core/renderer/command/resample/upsample.cpp

@@ -19,26 +19,26 @@ namespace AudioCore::AudioRenderer {
 static void SrcProcessFrame(std::span<s32> output, std::span<const s32> input,
                             const u32 target_sample_count, const u32 source_sample_count,
                             UpsamplerState* state) {
-    constexpr u32 WindowSize = 10;
-    constexpr std::array<Common::FixedPoint<24, 8>, WindowSize> SincWindow1{
-        51.93359375f, -18.80078125f, 9.73046875f, -5.33203125f, 2.84375f,
-        -1.41015625f, 0.62109375f,   -0.2265625f, 0.0625f,      -0.00390625f,
+    static constexpr u32 WindowSize = 10;
+    static constexpr std::array<Common::FixedPoint<17, 15>, WindowSize> WindowedSinc1{
+        0.95376587f,   -0.12872314f, 0.060028076f,  -0.032470703f, 0.017669678f,
+        -0.009124756f, 0.004272461f, -0.001739502f, 0.000579834f,  -0.000091552734f,
     };
-    constexpr std::array<Common::FixedPoint<24, 8>, WindowSize> SincWindow2{
-        105.35546875f, -24.52734375f, 11.9609375f,  -6.515625f, 3.52734375f,
-        -1.796875f,    0.828125f,     -0.32421875f, 0.1015625f, -0.015625f,
+    static constexpr std::array<Common::FixedPoint<17, 15>, WindowSize> WindowedSinc2{
+        0.8230896f,    -0.19161987f,  0.093444824f,  -0.05090332f,   0.027557373f,
+        -0.014038086f, 0.0064697266f, -0.002532959f, 0.00079345703f, -0.00012207031f,
     };
-    constexpr std::array<Common::FixedPoint<24, 8>, WindowSize> SincWindow3{
-        122.08203125f, -16.47656250f, 7.68359375f,  -4.15625000f, 2.26171875f,
-        -1.16796875f,  0.54687500f,   -0.22265625f, 0.07421875f,  -0.01171875f,
+    static constexpr std::array<Common::FixedPoint<17, 15>, WindowSize> WindowedSinc3{
+        0.6298828f,    -0.19274902f, 0.09725952f,    -0.05319214f,  0.028625488f,
+        -0.014373779f, 0.006500244f, -0.0024719238f, 0.0007324219f, -0.000091552734f,
     };
-    constexpr std::array<Common::FixedPoint<24, 8>, WindowSize> SincWindow4{
-        23.73437500f, -9.62109375f, 5.07812500f,  -2.78125000f, 1.46875000f,
-        -0.71484375f, 0.30859375f,  -0.10546875f, 0.02734375f,  0.00000000f,
+    static constexpr std::array<Common::FixedPoint<17, 15>, WindowSize> WindowedSinc4{
+        0.4057312f,    -0.1468811f,  0.07601929f,    -0.041656494f,  0.022216797f,
+        -0.011016846f, 0.004852295f, -0.0017700195f, 0.00048828125f, -0.000030517578f,
     };
-    constexpr std::array<Common::FixedPoint<24, 8>, WindowSize> SincWindow5{
-        80.62500000f, -24.67187500f, 12.44921875f, -6.80859375f, 3.66406250f,
-        -1.83984375f, 0.83203125f,   -0.31640625f, 0.09375000f,  -0.01171875f,
+    static constexpr std::array<Common::FixedPoint<17, 15>, WindowSize> WindowedSinc5{
+        0.1854248f,    -0.075164795f, 0.03967285f,    -0.021728516f,  0.011474609f,
+        -0.005584717f, 0.0024108887f, -0.0008239746f, 0.00021362305f, 0.0f,
     };
 
     if (!state->initialized) {
@@ -91,52 +91,31 @@ static void SrcProcessFrame(std::span<s32> output, std::span<const s32> input,
             static_cast<u16>((state->history_output_index + 1) % UpsamplerState::HistorySize);
     };
 
-    auto calculate_sample = [&state](std::span<const Common::FixedPoint<24, 8>> coeffs1,
-                                     std::span<const Common::FixedPoint<24, 8>> coeffs2) -> s32 {
+    auto calculate_sample = [&state](std::span<const Common::FixedPoint<17, 15>> coeffs1,
+                                     std::span<const Common::FixedPoint<17, 15>> coeffs2) -> s32 {
         auto output_index{state->history_output_index};
-        auto start_pos{output_index - state->history_start_index + 1U};
-        auto end_pos{10U};
+        u64 result{0};
 
-        if (start_pos < 10) {
-            end_pos = start_pos;
-        }
+        for (u32 coeff_index = 0; coeff_index < 10; coeff_index++) {
+            result += static_cast<u64>(state->history[output_index].to_raw()) *
+                      coeffs1[coeff_index].to_raw();
 
-        u64 prev_contrib{0};
-        u32 coeff_index{0};
-        for (; coeff_index < end_pos; coeff_index++, output_index--) {
-            prev_contrib += static_cast<u64>(state->history[output_index].to_raw()) *
-                            coeffs1[coeff_index].to_raw();
-        }
-
-        auto end_index{state->history_end_index};
-        for (; start_pos < 9; start_pos++, coeff_index++, end_index--) {
-            prev_contrib += static_cast<u64>(state->history[end_index].to_raw()) *
-                            coeffs1[coeff_index].to_raw();
+            output_index = output_index == state->history_start_index ? state->history_end_index
+                                                                      : output_index - 1;
         }
 
         output_index =
             static_cast<u16>((state->history_output_index + 1) % UpsamplerState::HistorySize);
-        start_pos = state->history_end_index - output_index + 1U;
-        end_pos = 10U;
 
-        if (start_pos < 10) {
-            end_pos = start_pos;
+        for (u32 coeff_index = 0; coeff_index < 10; coeff_index++) {
+            result += static_cast<u64>(state->history[output_index].to_raw()) *
+                      coeffs2[coeff_index].to_raw();
+
+            output_index = output_index == state->history_end_index ? state->history_start_index
+                                                                    : output_index + 1;
         }
 
-        u64 next_contrib{0};
-        coeff_index = 0;
-        for (; coeff_index < end_pos; coeff_index++, output_index++) {
-            next_contrib += static_cast<u64>(state->history[output_index].to_raw()) *
-                            coeffs2[coeff_index].to_raw();
-        }
-
-        auto start_index{state->history_start_index};
-        for (; start_pos < 9; start_pos++, start_index++, coeff_index++) {
-            next_contrib += static_cast<u64>(state->history[start_index].to_raw()) *
-                            coeffs2[coeff_index].to_raw();
-        }
-
-        return static_cast<s32>(((prev_contrib >> 15) + (next_contrib >> 15)) >> 8);
+        return static_cast<s32>(result >> (8 + 15));
     };
 
     switch (state->ratio.to_int_floor()) {
@@ -150,23 +129,23 @@ static void SrcProcessFrame(std::span<s32> output, std::span<const s32> input,
                 break;
 
             case 1:
-                output[write_index] = calculate_sample(SincWindow3, SincWindow4);
+                output[write_index] = calculate_sample(WindowedSinc1, WindowedSinc5);
                 break;
 
             case 2:
-                output[write_index] = calculate_sample(SincWindow2, SincWindow1);
+                output[write_index] = calculate_sample(WindowedSinc2, WindowedSinc4);
                 break;
 
             case 3:
-                output[write_index] = calculate_sample(SincWindow5, SincWindow5);
+                output[write_index] = calculate_sample(WindowedSinc3, WindowedSinc3);
                 break;
 
             case 4:
-                output[write_index] = calculate_sample(SincWindow1, SincWindow2);
+                output[write_index] = calculate_sample(WindowedSinc4, WindowedSinc2);
                 break;
 
             case 5:
-                output[write_index] = calculate_sample(SincWindow4, SincWindow3);
+                output[write_index] = calculate_sample(WindowedSinc5, WindowedSinc1);
                 break;
             }
             state->sample_index = static_cast<u8>((state->sample_index + 1) % 6);
@@ -183,11 +162,11 @@ static void SrcProcessFrame(std::span<s32> output, std::span<const s32> input,
                 break;
 
             case 1:
-                output[write_index] = calculate_sample(SincWindow2, SincWindow1);
+                output[write_index] = calculate_sample(WindowedSinc2, WindowedSinc4);
                 break;
 
             case 2:
-                output[write_index] = calculate_sample(SincWindow1, SincWindow2);
+                output[write_index] = calculate_sample(WindowedSinc4, WindowedSinc2);
                 break;
             }
             state->sample_index = static_cast<u8>((state->sample_index + 1) % 3);
@@ -204,12 +183,12 @@ static void SrcProcessFrame(std::span<s32> output, std::span<const s32> input,
                 break;
 
             case 1:
-                output[write_index] = calculate_sample(SincWindow1, SincWindow2);
+                output[write_index] = calculate_sample(WindowedSinc4, WindowedSinc2);
                 break;
 
             case 2:
                 increment();
-                output[write_index] = calculate_sample(SincWindow2, SincWindow1);
+                output[write_index] = calculate_sample(WindowedSinc2, WindowedSinc4);
                 break;
             }
             state->sample_index = static_cast<u8>((state->sample_index + 1) % 3);

src/audio_core/renderer/effect/effect_info_base.h

@@ -192,7 +192,7 @@ public:
     /**
      * Get this effect's parameter data.
      *
-     * @return Pointer to the parametter, must be cast to the correct type.
+     * @return Pointer to the parameter, must be cast to the correct type.
      */
     u8* GetParameter() {
         return parameter.data();
@@ -201,7 +201,7 @@ public:
     /**
      * Get this effect's parameter data.
      *
-     * @return Pointer to the parametter, must be cast to the correct type.
+     * @return Pointer to the parameter, must be cast to the correct type.
      */
     u8* GetStateBuffer() {
         return state.data();

src/audio_core/renderer/effect/i3dl2.h

@@ -104,7 +104,8 @@ public:
         }
 
         void Write(const Common::FixedPoint<50, 14> sample) {
-            *(input++) = sample;
+            *input = sample;
+            input++;
             if (input >= buffer_end) {
                 input = buffer.data();
             }

src/audio_core/renderer/effect/reverb.h

@@ -79,12 +79,10 @@ public:
                 return;
             }
             sample_count = delay_time;
-            input = &buffer[(output - buffer.data() + sample_count) % (sample_count_max + 1)];
+            input = &buffer[0];
         }
 
         Common::FixedPoint<50, 14> Tick(const Common::FixedPoint<50, 14> sample) {
-            Write(sample);
-
             auto out_sample{Read()};
 
             output++;
@@ -92,6 +90,7 @@ public:
                 output = buffer.data();
             }
 
+            Write(sample);
             return out_sample;
         }
 
@@ -100,7 +99,8 @@ public:
         }
 
         void Write(const Common::FixedPoint<50, 14> sample) {
-            *(input++) = sample;
+            *input = sample;
+            input++;
             if (input >= buffer_end) {
                 input = buffer.data();
             }

src/audio_core/renderer/memory/memory_pool_info.h

@@ -29,7 +29,7 @@ public:
      */
     enum class State {
         Invalid,
-        Aquired,
+        Acquired,
         RequestDetach,
         Detached,
         RequestAttach,

src/audio_core/renderer/memory/pool_mapper.cpp

@@ -92,7 +92,7 @@ bool PoolMapper::TryAttachBuffer(BehaviorInfo::ErrorInfo& error_info, AddressInf
     address_info.Setup(address, size);
 
     if (!FillDspAddr(address_info)) {
-        error_info.error_code = Service::Audio::ERR_POOL_MAPPING_FAILED;
+        error_info.error_code = Service::Audio::ResultInvalidAddressInfo;
         error_info.address = address;
         return force_map;
     }

src/audio_core/renderer/mix/mix_context.h

@@ -93,7 +93,7 @@ public:
      * Splitter sort, traverse the splitter node graph and sort the sorted mixes from results.
      *
      * @param splitter_context - Splitter context for the sort.
-     * @return True if the sort was successful, othewise false.
+     * @return True if the sort was successful, otherwise false.
      */
     bool TSortInfo(const SplitterContext& splitter_context);
 

src/audio_core/renderer/performance/performance_detail.h

@@ -33,7 +33,7 @@ struct PerformanceDetailVersion1 {
     /* 0x0D */ PerformanceEntryType entry_type;
 };
 static_assert(sizeof(PerformanceDetailVersion1) == 0x10,
-              "PerformanceDetailVersion1 has the worng size!");
+              "PerformanceDetailVersion1 has the wrong size!");
 
 struct PerformanceDetailVersion2 {
     /* 0x00 */ u32 node_id;
@@ -45,6 +45,6 @@ struct PerformanceDetailVersion2 {
     /* 0x14 */ char unk14[0x4];
 };
 static_assert(sizeof(PerformanceDetailVersion2) == 0x18,
-              "PerformanceDetailVersion2 has the worng size!");
+              "PerformanceDetailVersion2 has the wrong size!");
 
 } // namespace AudioCore::AudioRenderer

src/audio_core/renderer/performance/performance_entry.h

@@ -22,7 +22,7 @@ struct PerformanceEntryVersion1 {
     /* 0x0C */ PerformanceEntryType entry_type;
 };
 static_assert(sizeof(PerformanceEntryVersion1) == 0x10,
-              "PerformanceEntryVersion1 has the worng size!");
+              "PerformanceEntryVersion1 has the wrong size!");
 
 struct PerformanceEntryVersion2 {
     /* 0x00 */ u32 node_id;
@@ -32,6 +32,6 @@ struct PerformanceEntryVersion2 {
     /* 0x0D */ char unk0D[0xB];
 };
 static_assert(sizeof(PerformanceEntryVersion2) == 0x18,
-              "PerformanceEntryVersion2 has the worng size!");
+              "PerformanceEntryVersion2 has the wrong size!");
 
 } // namespace AudioCore::AudioRenderer

src/audio_core/renderer/performance/performance_frame_header.h

@@ -16,7 +16,7 @@ struct PerformanceFrameHeaderVersion1 {
     /* 0x14 */ u32 frame_index;
 };
 static_assert(sizeof(PerformanceFrameHeaderVersion1) == 0x18,
-              "PerformanceFrameHeaderVersion1 has the worng size!");
+              "PerformanceFrameHeaderVersion1 has the wrong size!");
 
 struct PerformanceFrameHeaderVersion2 {
     /* 0x00 */ u32 magic; // "PERF"
@@ -31,6 +31,6 @@ struct PerformanceFrameHeaderVersion2 {
     /* 0x25 */ char unk25[0xB];
 };
 static_assert(sizeof(PerformanceFrameHeaderVersion2) == 0x30,
-              "PerformanceFrameHeaderVersion2 has the worng size!");
+              "PerformanceFrameHeaderVersion2 has the wrong size!");
 
 } // namespace AudioCore::AudioRenderer