Move to C++20 (#2088)

* update circleci
* update appveyor android builds
* update appveyor to vs2019
* build with c++20
* fix for vs2019 breakage
* update circleci meson for macos

Author: borg323

.circleci/config.yml

@@ -2,26 +2,41 @@ version: 2
 jobs:
   build:
     docker:
-      - image: danieluranga/leela_chess_zero-lc0_ubuntu_builder:0.0.8
+      - image: nvidia/cuda:11.6.1-cudnn8-devel-ubuntu20.04
     steps:
       - checkout
+      - run:
+          name: Install build tools
+          command: |
+            apt-get update
+            apt-get -y install git python3-pip gcc-10 g++-10 clang-12 zlib1g zlib1g-dev
+            pip3 install meson==0.63
+            pip3 install ninja
       - run:
           name: "Pull Submodules"
           command: git submodule update --init
-      - run:
-          name: Update Meson
-          command: pip3 install --upgrade meson==0.58.1
       - run:
           name: Meson GCC
           environment:
-            CC: gcc-8
-            CXX: g++-8
+            CC: gcc-10
+            CXX: g++-10
           command: meson build-gcc -Dgtest=false
+      - run:
+          name: Meson Clang
+          environment:
+            CC: clang-12
+            CXX: clang++-12
+          command: meson build-clang -Dgtest=false
       - run:
           name: Build GCC
           command: |
             cd build-gcc
             ninja -j 4
+      - run:
+          name: Build Clang
+          command: |
+            cd build-clang
+            ninja -j 4
   "mac":
     macos:
       xcode: 14.1.0
@@ -34,7 +49,7 @@ jobs:
       - run:
           name: Install build tools
           command: |
-            pip3 install meson==0.63
+            pip3 install meson==1.3.0
             pip3 install ninja
             curl -LJOs https://github.com/ispc/ispc/releases/download/v1.21.0/ispc-v1.21.0-macOS.universal.tar.gz
             tar xf ispc-v1.21.0-macOS.universal.tar.gz

appveyor.yml

@@ -2,7 +2,7 @@ version: '{build}'
 configuration: Release
 platform: x64
 image:
-- Visual Studio 2017
+- Visual Studio 2019
 environment:
   matrix:
   - NAME: gpu-nvidia-cudnn
@@ -47,7 +47,7 @@ install:
 - cmd: set NET_HASH=3e3444370b9fe413244fdc79671a490e19b93d3cca1669710ffeac890493d198
 - cmd: IF NOT %OPENCL%==true IF NOT %DX%==true set NET=791556
 - cmd: IF NOT %OPENCL%==true IF NOT %DX%==true set NET_HASH=f404e156ceb2882470fd8c032b8754af0fa0b71168328912eaef14671a256e34
-- cmd: call "C:\Program Files (x86)\Microsoft Visual Studio\2017\Community\VC\Auxiliary\Build\vcvarsall.bat" amd64
+- cmd: call "C:\Program Files (x86)\Microsoft Visual Studio\2019\Community\VC\Auxiliary\Build\vcvarsall.bat" amd64
 - cmd: set DNNL_NAME=dnnl_win_1.5.0_cpu_vcomp
 - cmd: IF %NAME%==cpu-dnnl IF NOT EXIST C:\cache\%DNNL_NAME% appveyor DownloadFile https://github.com/oneapi-src/oneDNN/releases/download/v1.5/dnnl_win_1.5.0_cpu_vcomp.zip
 - cmd: IF %NAME%==cpu-dnnl IF NOT EXIST C:\cache\%DNNL_NAME% 7z x dnnl_win_1.5.0_cpu_vcomp.zip -oC:\cache
@@ -65,26 +65,26 @@ install:
 - cmd: IF %ISPC%==true IF NOT EXIST C:\cache\ispc-v1.13.0-windows appveyor DownloadFile https://github.com/ispc/ispc/releases/download/v1.13.0/ispc-v1.13.0-windows.zip
 - cmd: IF %ISPC%==true IF NOT EXIST C:\cache\ispc-v1.13.0-windows 7z x ispc-v1.13.0-windows.zip -oC:\cache\ispc-v1.13.0-windows
 - cmd: IF %ISPC%==true set PATH=C:\cache\ispc-v1.13.0-windows\bin;%PATH%
-- cmd: set "CUDA_PATH=C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v10.0"
+- cmd: set "CUDA_PATH=C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v10.1"
 - cmd: IF %CUDNN%==true IF NOT EXIST "%CUDA_PATH%\cuda" set CUDNN_INSTALL=1
-- cmd: IF DEFINED CUDNN_INSTALL appveyor DownloadFile https://developer.nvidia.com/compute/cuda/10.0/Prod/network_installers/cuda_10.0.130_win10_network
-- cmd: IF DEFINED CUDNN_INSTALL cuda_10.0.130_win10_network -s nvcc_10.0 cublas_dev_10.0 cublas_10.0 cudart_10.0
-- cmd: IF DEFINED CUDNN_INSTALL appveyor DownloadFile http://developer.download.nvidia.com/compute/redist/cudnn/v7.4.2/cudnn-10.0-windows10-x64-v7.4.2.24.zip
-- cmd: IF DEFINED CUDNN_INSTALL 7z x cudnn-10.0-windows10-x64-v7.4.2.24.zip -o"%CUDA_PATH%"
+- cmd: IF DEFINED CUDNN_INSTALL appveyor DownloadFile https://developer.download.nvidia.com/compute/cuda/10.1/Prod/network_installers/cuda_10.1.243_win10_network.exe
+- cmd: IF DEFINED CUDNN_INSTALL cuda_10.1.243_win10_network -s nvcc_10.1 cublas_dev_10.1 cublas_10.1 cudart_10.1
+- cmd: IF DEFINED CUDNN_INSTALL appveyor DownloadFile https://developer.download.nvidia.com/compute/redist/cudnn/v7.5.1/cudnn-10.1-windows10-x64-v7.5.1.10.zip
+- cmd: IF DEFINED CUDNN_INSTALL 7z x cudnn-10.1-windows10-x64-v7.5.1.10.zip -o"%CUDA_PATH%"
 - cmd: IF %CUDNN%==false set "CUDA_PATH=C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.1"
 - cmd: IF %CUDA%==true IF NOT EXIST "%CUDA_PATH%" set CUDA_INSTALL=1
 - cmd: IF DEFINED CUDA_INSTALL appveyor DownloadFile https://developer.download.nvidia.com/compute/cuda/11.1.0/network_installers/cuda_11.1.0_win10_network.exe
 - cmd: IF DEFINED CUDA_INSTALL cuda_11.1.0_win10_network.exe -s nvcc_11.1 cublas_dev_11.1 cublas_11.1 cudart_11.1 documentation_11.1
 - cmd: IF %CUDA%==true set PATH=%CUDA_PATH%\bin;%PATH%
-- cmd: set PATH=C:\Python36;C:\Python36\scripts;%PATH%
-- cmd: pip3 install --upgrade meson==0.55.3
-- cmd: set MIMALLOC_PATH=C:\cache\mimalloc-1.7.1
-- cmd: IF %ANDROID%==false IF NOT EXIST "%MIMALLOC_PATH%" appveyor DownloadFile https://github.com/microsoft/mimalloc/archive/refs/tags/v1.7.1.zip
-- cmd: IF %ANDROID%==false IF NOT EXIST "%MIMALLOC_PATH%" 7z x v1.7.1.zip -oC:\cache\
-- cmd: IF %ANDROID%==false IF NOT EXIST "%MIMALLOC_PATH%"\out msbuild "%MIMALLOC_PATH%"\ide\vs2017\mimalloc-override.vcxproj /p:Configuration=Release /m
-- cmd: IF %NAME%==android IF NOT EXIST C:\ndk\android-ndk-r19c\toolchains\llvm\prebuilt\windows-x86_64 appveyor DownloadFile https://dl.google.com/android/repository/android-ndk-r19c-windows-x86_64.zip
-- cmd: IF %NAME%==android IF NOT EXIST C:\ndk\android-ndk-r19c\toolchains\llvm\prebuilt\windows-x86_64 7z x android-ndk-r19c-windows-x86_64.zip -oC:\ndk
-- cmd: IF %NAME%==android set PATH=C:\ndk\android-ndk-r19c\toolchains\llvm\prebuilt\windows-x86_64\bin;%PATH%
+- cmd: set PATH=C:\Python310;C:\Python310\scripts;%PATH%
+#- cmd: pip3 install --upgrade meson==0.55.3
+- cmd: set MIMALLOC_PATH=C:\cache\mimalloc-1.8.7
+- cmd: IF %ANDROID%==false IF NOT EXIST "%MIMALLOC_PATH%" appveyor DownloadFile https://github.com/microsoft/mimalloc/archive/refs/tags/v1.8.7.zip
+- cmd: IF %ANDROID%==false IF NOT EXIST "%MIMALLOC_PATH%" 7z x v1.8.7.zip -oC:\cache\
+- cmd: IF %ANDROID%==false IF NOT EXIST "%MIMALLOC_PATH%"\out msbuild "%MIMALLOC_PATH%"\ide\vs2019\mimalloc-override.vcxproj /p:Configuration=Release /m
+- cmd: IF %NAME%==android IF NOT EXIST C:\ndk\android-ndk-r27c\toolchains\llvm\prebuilt\windows-x86_64 appveyor DownloadFile https://dl.google.com/android/repository/android-ndk-r27c-windows.zip
+- cmd: IF %NAME%==android IF NOT EXIST C:\ndk\android-ndk-r27c\toolchains\llvm\prebuilt\windows-x86_64 7z x android-ndk-r27c-windows.zip -oC:\ndk
+- cmd: IF %NAME%==android set PATH=C:\ndk\android-ndk-r27c\toolchains\llvm\prebuilt\windows-x86_64\bin;%PATH%
 - cmd: IF %NAME%==android sed "s/clang+*/&.cmd/" cross-files/aarch64-linux-android >crossfile-aarch64
 - cmd: IF %NAME%==android IF NOT EXIST C:\cache\OpenBLAS\android-aarch64 appveyor DownloadFile https://github.com/borg323/OpenBLAS/releases/download/android-0.3.27/openblas-android-aarch64.zip
 - cmd: IF %NAME%==android IF NOT EXIST C:\cache\OpenBLAS\android-aarch64 7z x openblas-android-aarch64.zip -oC:\cache\OpenBLAS
@@ -103,10 +103,10 @@ install:
 - cmd: cd C:\projects\lc0
 cache:
   - C:\cache
-  - 'C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v10.0'
+  - 'C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v10.1'
   - 'C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v11.1'
   - C:\projects\lc0\subprojects\packagecache
-  - C:\ndk\android-ndk-r19c\toolchains\llvm\prebuilt\windows-x86_64
+  - C:\ndk\android-ndk-r27c\toolchains\llvm\prebuilt\windows-x86_64
 before_build:
 - cmd: git submodule update --init --recursive
 - cmd: IF %BLAS%==true (echo.#define DEFAULT_MAX_PREFETCH 0 & echo.#define DEFAULT_TASK_WORKERS 0) > params_override.h
@@ -124,7 +124,7 @@ before_build:
 - cmd: SET EXTRA=
 - cmd: IF %ANDROID%==false SET EXTRA=-Db_vscrt=md
 - cmd: IF %ONNX_DML%==true SET EXTRA=-Db_vscrt=md -Donnx_libdir=C:\cache\%ONNX_NAME%\lib -Donnx_include=C:\cache\%ONNX_NAME%\include
-- cmd: IF %ANDROID%==false meson build --backend vs2017 --buildtype release -Dgtest=%GTEST% -Dopencl=%OPENCL% -Dblas=%BUILD_BLAS% -Ddnnl=true -Ddx=%DX% -Dcudnn=%CUDNN% -Donednn=%ONEDNN% -Dispc_native_only=false -Dnative_cuda=false -Dpopcnt=%POPCNT% -Df16c=%F16C% -Dcudnn_include="%CUDA_PATH%\include","%CUDA_PATH%\cuda\include" -Dcudnn_libdirs="%CUDA_PATH%\lib\x64","%CUDA_PATH%\cuda\lib\x64" -Dopenblas_include="%PKG_FOLDER%\OpenBLAS\dist64\include" -Dopenblas_libdirs="%PKG_FOLDER%\OpenBLAS\dist64\lib" -Ddnnl_dir="%PKG_FOLDER%\%DNNL_NAME%" -Dopencl_include="%PKG_FOLDER%\opencl-nug.0.777.77\build\native\include" -Dopencl_libdirs="%PKG_FOLDER%\opencl-nug.0.777.77\build\native\lib\x64" -Ddefault_library=static -Dmalloc=mimalloc -Dmimalloc_libdir="%MIMALLOC_PATH%"\out\msvc-x64\Release %EXTRA%
+- cmd: IF %ANDROID%==false meson build --backend vs2019 --buildtype release -Dgtest=%GTEST% -Dopencl=%OPENCL% -Dblas=%BUILD_BLAS% -Ddnnl=true -Ddx=%DX% -Dcudnn=%CUDNN% -Donednn=%ONEDNN% -Dispc_native_only=false -Dnative_cuda=false -Dpopcnt=%POPCNT% -Df16c=%F16C% -Dcudnn_include="%CUDA_PATH%\include","%CUDA_PATH%\cuda\include" -Dcudnn_libdirs="%CUDA_PATH%\lib\x64","%CUDA_PATH%\cuda\lib\x64" -Dopenblas_include="%PKG_FOLDER%\OpenBLAS\dist64\include" -Dopenblas_libdirs="%PKG_FOLDER%\OpenBLAS\dist64\lib" -Ddnnl_dir="%PKG_FOLDER%\%DNNL_NAME%" -Dopencl_include="%PKG_FOLDER%\opencl-nug.0.777.77\build\native\include" -Dopencl_libdirs="%PKG_FOLDER%\opencl-nug.0.777.77\build\native\lib\x64" -Ddefault_library=static -Dmalloc=mimalloc -Dmimalloc_libdir="%MIMALLOC_PATH%"\out\msvc-x64\Release %EXTRA%
 - cmd: IF %ANDROID%==true meson arm64-v8a --buildtype release -Dgtest=false -Dopenblas_include="%PKG_FOLDER%\OpenBLAS\android-aarch64\include" -Dopenblas_libdirs="%PKG_FOLDER%\OpenBLAS\android-aarch64\lib" -Dembed=%EMBED% -Ddefault_library=static --cross-file crossfile-aarch64
 - cmd: IF %ANDROID%==true meson armeabi-v7a --buildtype release -Dgtest=false -Dopenblas_include="%PKG_FOLDER%\OpenBLAS\android-armv7a\include" -Dopenblas_libdirs="%PKG_FOLDER%\OpenBLAS\android-armv7a\lib" -Dembed=%EMBED% -Ddefault_library=static --cross-file crossfile-armv7a -Dispc=false -Dneon=false
 build_script:

cross-files/aarch64-linux-android

@@ -1,5 +1,5 @@
 
-# Tested with Android NDK r19c, default toolchain
+# Tested with Android NDK r27c, default toolchain
 # Targeting API level 21
 
 # Set the toolchain path on your environment
@@ -17,8 +17,8 @@ cpp_link_args = ['-llog', '-static-libstdc++']
 [binaries]
 c = 'aarch64-linux-android21-clang'
 cpp = 'aarch64-linux-android21-clang++'
-ar = 'aarch64-linux-android-ar'
-strip = 'aarch64-linux-android-strip'
-ld = 'aarch64-linux-android-ld'
-ranlib = 'aarch64-linux-android-ranlib'
-as = 'aarch64-linux-android-as'
+ar = 'llvm-ar'
+strip = 'llvm-strip'
+ld = 'ld'
+ranlib = 'llvm-ranlib'
+as = 'aarch64-linux-android21-clang'

cross-files/armv7a-linux-android

@@ -1,5 +1,5 @@
 
-# Tested with Android NDK r19c, default toolchain
+# Tested with Android NDK r27c, default toolchain
 # Targeting API level 21
 
 # When targeting API levels < 24 the build fails unless _FILE_OFFSET_BITS is unset.
@@ -24,8 +24,8 @@ cpp_link_args = ['-llog', '-static-libstdc++']
 [binaries]
 c = 'armv7a-linux-androideabi21-clang'
 cpp = 'armv7a-linux-androideabi21-clang++'
-ar = 'arm-linux-androideabi-ar'
-strip = 'arm-linux-androideabi-strip'
-ld = 'arm-linux-androideabi-ld'
-ranlib = 'arm-linux-androideabi-ranlib'
-as = 'arm-linux-androideabi-as'
+ar = 'llvm-ar'
+strip = 'llvm-strip'
+ld = 'ld'
+ranlib = 'llvm-ranlib'
+as = 'armv7a-linux-androideabi21-clang'

meson.build

@@ -15,7 +15,7 @@
 # along with Leela Chess.  If not, see <http://www.gnu.org/licenses/>.
 
 project('lc0', 'cpp',
-        default_options : ['cpp_std=c++17', 'b_ndebug=if-release', 'warning_level=3', 'b_lto=true', 'b_vscrt=mt'],
+        default_options : ['cpp_std=c++20', 'b_ndebug=if-release', 'warning_level=3', 'b_lto=true', 'b_vscrt=mt'],
         meson_version: '>=0.55')
 
 cc = meson.get_compiler('cpp')

scripts/appveyor_android_build.cmd

@@ -1,7 +1,7 @@
 cd arm64-v8a
 ninja
-aarch64-linux-android-strip lc0
+llvm-strip lc0
 cd C:\projects\lc0
 cd armeabi-v7a
 ninja
-arm-linux-androideabi-strip lc0
+llvm-strip lc0

scripts/appveyor_win_package.cmd

@@ -15,7 +15,7 @@ IF %NAME%==cpu-openblas 7z a lc0-%APPVEYOR_REPO_TAG_NAME%-windows-%NAME%.zip C:\
 IF %NAME%==cpu-dnnl 7z a lc0-%APPVEYOR_REPO_TAG_NAME%-windows-%NAME%.zip C:\cache\%DNNL_NAME%\bin\dnnl.dll
 IF %NAME%==onednn 7z a lc0-%APPVEYOR_REPO_TAG_NAME%-windows-%NAME%.zip C:\cache\%DNNL_NAME%\bin\dnnl.dll
 IF %OPENCL%==true 7z a lc0-%APPVEYOR_REPO_TAG_NAME%-windows-%NAME%.zip C:\cache\opencl-nug.0.777.77\build\native\bin\OpenCL.dll
-IF %CUDNN%==true 7z a lc0-%APPVEYOR_REPO_TAG_NAME%-windows-%NAME%.zip "%CUDA_PATH%\bin\cudart64_100.dll" "%CUDA_PATH%\bin\cublas64_100.dll"
+IF %CUDNN%==true 7z a lc0-%APPVEYOR_REPO_TAG_NAME%-windows-%NAME%.zip "%CUDA_PATH%\bin\cudart64_101.dll" "%CUDA_PATH%\bin\cublas64_10.dll" "%CUDA_PATH%\bin\cublasLt64_10.dll"
 IF %CUDNN%==true 7z a lc0-%APPVEYOR_REPO_TAG_NAME%-windows-%NAME%.zip "%CUDA_PATH%\cuda\bin\cudnn64_7.dll"
 IF %CUDA%==true 7z a lc0-%APPVEYOR_REPO_TAG_NAME%-windows-%NAME%.zip "%CUDA_PATH%\bin\cudart64_110.dll" "%CUDA_PATH%\bin\cublas64_11.dll" "%CUDA_PATH%\bin\cublasLt64_11.dll"
 IF %NAME%==cpu-dnnl copy "%PKG_FOLDER%\%DNNL_NAME%\LICENSE" dist\DNNL-LICENSE

src/neural/xla/onnx2hlo.cc

@@ -273,7 +273,7 @@ pblczero::XlaLiteralProto ConstOpMax(const pblczero::XlaLiteralProto& lhs,
             typename std::remove_reference<decltype(lhs)>::type::value_type;
         std::transform(lhs.begin(), lhs.end(), rhs.begin(),
                        std::back_inserter(*dst),
-                       [](T a, T b) { return std::max(a, b); });
+                       [](const T &a, const T &b) { return std::max(a, b); });
       });
   return result;
 }