Template repository to build PyTorch projects from source on any version of PyTorch/CUDA/cuDNN.

Overview

A Universal PyTorch Source-Build Docker Template

Template repository to build PyTorch from source on any version of PyTorch/CUDA/cuDNN.

PyTorch built from source is much faster (as much as x4 times on some benchmarks) than PyTorch installed from pip/conda but building from source is a difficult and bug-prone process.

This repository is a highly modular template to build any version of PyTorch from source on any version of CUDA. It provides an easy-to-use Dockerfile which can be integrated into any Linux-based image or project.

For researchers unfamiliar with Docker, the generated wheel files can be extracted to install PyTorch on their local environments.

Windows users may also use this project via WSL. See instructions below.

A Makefile is provided both as an interface for easy use and as a tutorial for building custom images.

Quick Start

Users are free to customize the train stage of the Dockerfile as they please. However, do not change the build stages unless absolutely necessary.

This project is a template, and users are expected to customize it to fit their needs.

The code is assumed to be running on a Linux host with the necessary NVIDIA Drivers and a recent version of Docker pre-installed. If this is not the case, install these first.

To build a training image, first edit the Dockerfile train stage to include desired packages from apt/conda/pip.

Then, visit https://developer.nvidia.com/cuda-gpus to find the Compute Capability (CC) of the target GPU device.

Finally, run make all CC=TARGET_CC(s).

Examples: (1) make all CC="8.6" for RTX 3090, (2) make all CC="7.5; 8.6" for both RTX 2080Ti and RTX 3090 (building for many GPU CCs will increase build time).

This will result in an image, pytorch_source:train, which can be used for training.

Note that CCs for devices not available during the build can be used to build the image.

For example, if the image must be used on an RTX 2080Ti machine but the user only has an RTX 3090, the user can set CC="7.5" to enable the image to operate on the RTX 2080Ti GPU.

See https://pytorch.org/docs/stable/cpp_extension.html for an in-depth guide on how to set TORCH_CUDA_ARCH_LIST, which is specified by CC in the Makefile.

Even if you do not wish to use Docker directly in your project, you may still find this template useful.

The wheel files generated by the build can be used in any Python environment with no dependency on Docker.

This project can thus be used to generate custom wheel files, improving both training and inference speeds dramatically, for any desired environment (conda, pip, etc.).

Makefile Explanation

The Makefile is designed to make using this package simple and modular.

The first image to be created is pytorch_source:build_install, which contains all necessary packages for the build.

The second image is pytorch_source:build_torch-v1.9.1 (by default), which contains the wheels for PyTorch, TorchVision, TorchText, and TorchAudio with settings for PyTorch 1.9.1 on Ubuntu 20.04 LTS with Python 3.8, CUDA 11.2.2 and cuDNN 8.

The second image exists to cache the results of the build process.

If you do not wish to use Docker and would like to only extract the .whl wheel files for a pip install on your environment, the generated wheel files can be found in the /tmp/dist directory.

Saving the build results also allows more convenient version switching in case different PyTorch versions (different CUDA version, different library version, etc.) are needed.

The final image is pytorch_source:train, which is the image to be used for actual training.

It relies on the previous stages only for the build artifacts (wheels, etc.) and nothing else.

This makes it very simple to create different training images optimized for different environments and GPU devices.

Because PyTorch has already been built, the training image only needs to download the remaining apt/conda/pip packages. Moreover, caching is implemented to speed up even this process.

Timezone Settings

International users may find this section helpful.

The train image has its timezone set by the TZ variable using the tzdata package. The default timezone is Asia/Seoul but this can be changed by specifying the TZ variable when calling make. Use IANA time zone names to specify the desired timezone.

Example: make all CC="8.6" TZ=America/Los_Angeles to use LA time on the training image.

NOTE: Only the training image has timezone settings. The installation and build images do not use timezone information.

In addition, the training image has apt and pip installation URLs updated for Korean users. If you wish to speed up your installs, please find URLs optimized for your location, though the installation caches may make this unnecessary.

Specific PyTorch Version

To change the version of PyTorch, set the PYTORCH_VERSION_TAG, TORCHVISION_VERSION_TAG, TORCHTEXT_VERSION_TAG, and TORCHAUDIO_VERSION_TAG variables to matching versions.

The *_TAG variables must be GitHub tags or branch names of those repositories. Visit the GitHub repositories of each library to find the appropriate tags.

PyTorch subsidiary libraries only work with matching versions of PyTorch.

Example: To build on an RTX 3090 GPU with PyTorch 1.9.1, use the following command:

make all CC="8.6" PYTORCH_VERSION_TAG=v1.9.1 TORCHVISION_VERSION_TAG=v0.10.1 TORCHTEXT_VERSION_TAG=v0.10.1 TORCHAUDIO_VERSION_TAG=v0.9.1.

The resulting image, pytorch_source:train, can be used for training with PyTorch 1.9.1 on GPUs with Compute Capability 8.6.

Multiple Training Images

To use multiple training images on the same host, give a different name to TRAIN_NAME, which has a default value of train.

New training images can be created without having to rebuild PyTorch if the same build image is used for different training images. Creating new training images takes only a few minutes at most.

This is useful for the following use cases.

  1. Allowing different users, who have different UID/GIDs, to use separate training images.
  2. Using different versions of the final training image with different library installations and configurations.

For example, if pytorch_source:build_torch-v1.9.1 has already been built, Alice and Bob would use the following commands to create separate images.

Alice: make build-train CC="8.6" TORCH_NAME=build_torch-v1.9.1 PYTORCH_VERSION_TAG=v1.9.1 TORCHVISION_VERSION_TAG=v0.10.1 TORCHTEXT_VERSION_TAG=v0.10.1 TORCHAUDIO_VERSION_TAG=v0.9.1 TRAIN_NAME=train_alice

Bob: make build-train CC="8.6" TORCH_NAME=build_torch-v1.9.1 PYTORCH_VERSION_TAG=v1.9.1 TORCHVISION_VERSION_TAG=v0.10.1 TORCHTEXT_VERSION_TAG=v0.10.1 TORCHAUDIO_VERSION_TAG=v0.9.1 TRAIN_NAME=train_bob

This way, Alice's image would have her UID/GID while Bob's image would have his UID/GID. This procedure is necessary because training images have their users set during build. Also, different users may install different libraries in their training images. Their environment variables and other settings may also be different.

Word of Caution

When using build images such as pytorch_source:build_torch-v1.9.1 as a build cache for creating new training images, the user must re-specify all build arguments (variables specified by ARG and ENV using --build-arg) of all previous layers.

Otherwise, the default values for these arguments will be given to the Dockerfile and a cache miss will occur because of the different input values.

This will both waste time rebuilding previous layers and, more importantly, cause inconsistency in the training images due to environment mismatch.

The BUILDKIT_INLINE_CACHE must also be given to an image to use it as a cache later. See https://docs.docker.com/engine/reference/commandline/build/#specifying-external-cache-sources for more information.

Advanced Usage

The Makefile provides the *-full commands for advanced usage.

make all-full CC=YOUR_GPU_CC TRAIN_NAME=train_cu102 will create pytorch_source:build_torch-v1.9.1-ubuntu18.04-cuda10.2-cudnn8-py3.9 and pytorch_source:train_cu102 by default.

The example images can be used for training/deployment on CUDA 10 devices such as the GTX 1080Ti.

Also, *-clean commands are provided to check for reliance on caches from previous builds.

Specific CUDA Version

Set CUDA_VERSION, CUDNN_VERSION, and MAGMA_VERSION to change CUDA versions. PYTHON_VERSION may also be changed if necessary.

This will create a build image that can be used as a cache to create training images with the build-train command.

Also, the extensive use of caching in the project means that the second build is much faster than the first build. This may be advantageous if many images must be created for multiple PyTorch/CUDA versions.

Specific Linux Distro

CentOS and UBI images can be created with only minor edits to the Dockerfile. Read the Dockerfile for full instructions.

Set LINUX_DISTRO and DISTRO_VERSION arguments afterwards.

Windows

Windows users may use template by updating to Windows 11 and installing Windows Subsystem for Linux (WSL). WSL on Windows 11 gives a similar experience to using native Linux.

This project has been tested on WSL on Windows 11 with the WSL CUDA driver and Docker Desktop for Windows.

Comments
  • Error during build

    Error during build

    Hi,

    I cloned the github version on the main branch, and then executed the following command make all-full CC="8.0" TRAIN_NAME=train_cu102, but I got the following error:

    DOCKER_BUILDKIT=1 docker build \
    	--target build-install \
    	--tag pytorch_source:build_install-ubuntu18.04-cuda10.2-cudnn8-py3.9 \
    	--build-arg LINUX_DISTRO=ubuntu \
    	--build-arg DISTRO_VERSION=18.04 \
    	--build-arg CUDA_VERSION=10.2 \
    	--build-arg CUDNN_VERSION=8 \
    	--build-arg MAGMA_VERSION=102   \
    	--build-arg PYTHON_VERSION=3.9 \
    	--build-arg BUILDKIT_INLINE_CACHE=1 \
    	- < Dockerfile
    [+] Building 0.7s (21/21) FINISHED                                                                       
     => [internal] load build definition from Dockerfile                                                0.0s
     => => transferring dockerfile: 12.61kB                                                             0.0s
     => [internal] load .dockerignore                                                                   0.0s
     => => transferring context: 2B                                                                     0.0s
     => resolve image config for docker.io/docker/dockerfile:1.3.0-labs                                 0.4s
     => CACHED docker-image://docker.io/docker/dockerfile:1.3.0-labs@sha256:03ca0e50aa4b6e76365fa9a560  0.0s
     => [internal] load build definition from Dockerfile                                                0.0s
     => [internal] load .dockerignore                                                                   0.0s
     => [internal] load metadata for docker.io/nvidia/cuda:10.2-cudnn8-devel-ubuntu18.04                0.0s
     => [build-base-ubuntu 1/3] FROM docker.io/nvidia/cuda:10.2-cudnn8-devel-ubuntu18.04                0.0s
     => CACHED [build-base-ubuntu 2/3] RUN rm -f /etc/apt/apt.conf.d/docker-clean;     echo 'Binary::a  0.0s
     => CACHED [build-base-ubuntu 3/3] RUN --mount=type=cache,id=apt-cache-build,target=/var/cache/apt  0.0s
     => CACHED [build-base 1/3] RUN /usr/sbin/update-ccache-symlinks                                    0.0s
     => CACHED [build-base 2/3] RUN mkdir /opt/ccache && ccache --set-config=cache_dir=/opt/ccache &&   0.0s
     => CACHED [build-base 3/3] RUN curl -fsSL -v -o ~/miniconda.sh -O  https://repo.anaconda.com/mini  0.0s
     => CACHED [build-install 1/6] RUN --mount=type=cache,id=conda-build,target=/opt/conda/pkgs     co  0.0s
     => CACHED [build-install 2/6] WORKDIR /opt                                                         0.0s
     => CACHED [build-install 3/6] RUN git clone --recursive --jobs 0 https://github.com/pytorch/pytor  0.0s
     => CACHED [build-install 4/6] RUN git clone --recursive --jobs 0 https://github.com/pytorch/visio  0.0s
     => CACHED [build-install 5/6] RUN git clone --recursive --jobs 0 https://github.com/pytorch/text.  0.0s
     => CACHED [build-install 6/6] RUN git clone --recursive --jobs 0 https://github.com/pytorch/audio  0.0s
     => exporting to image                                                                              0.0s
     => => exporting layers                                                                             0.0s
     => => writing image sha256:81330a777d1f8a68c315ec6f1859c16ea71076ebd84a5c0bbd88fa6a333949b5        0.0s
     => => naming to docker.io/library/pytorch_source:build_install-ubuntu18.04-cuda10.2-cudnn8-py3.9   0.0s
     => exporting cache                                                                                 0.0s
     => => preparing build cache for export                                                             0.0s
    DOCKER_BUILDKIT=1 docker build \
    	--target train-builds \
    	--cache-from=pytorch_source:build_install-ubuntu18.04-cuda10.2-cudnn8-py3.9 \
    	--tag pytorch_source:build_torch-v1.9.1-ubuntu18.04-cuda10.2-cudnn8-py3.9 \
    	--build-arg TORCH_CUDA_ARCH_LIST="8.0" \
    	--build-arg PYTORCH_VERSION_TAG=v1.9.1 \
    	--build-arg TORCHVISION_VERSION_TAG=v0.10.1 \
    	--build-arg TORCHTEXT_VERSION_TAG=v0.10.1 \
    	--build-arg TORCHAUDIO_VERSION_TAG=v0.9.1 \
    	--build-arg LINUX_DISTRO=ubuntu \
    	--build-arg DISTRO_VERSION=18.04 \
    	--build-arg CUDA_VERSION=10.2 \
    	--build-arg CUDNN_VERSION=8 \
    	--build-arg MAGMA_VERSION=102   \
    	--build-arg PYTHON_VERSION=3.9 \
    	--build-arg BUILDKIT_INLINE_CACHE=1 \
    	- < Dockerfile
    [+] Building 29.1s (24/37)                                                                               
     => [internal] load build definition from Dockerfile                                                0.0s
     => => transferring dockerfile: 12.61kB                                                             0.0s
     => [internal] load .dockerignore                                                                   0.0s
     => => transferring context: 2B                                                                     0.0s
     => resolve image config for docker.io/docker/dockerfile:1.3.0-labs                                 0.5s
     => CACHED docker-image://docker.io/docker/dockerfile:1.3.0-labs@sha256:03ca0e50aa4b6e76365fa9a560  0.0s
     => [internal] load build definition from Dockerfile                                                0.0s
     => [internal] load .dockerignore                                                                   0.0s
     => [internal] load metadata for docker.io/nvidia/cuda:10.2-cudnn8-devel-ubuntu18.04                0.0s
     => importing cache manifest from pytorch_source:build_install-ubuntu18.04-cuda10.2-cudnn8-py3.9    0.0s
     => [train-builds 1/5] FROM docker.io/nvidia/cuda:10.2-cudnn8-devel-ubuntu18.04                     0.0s
     => CACHED [build-base-ubuntu 2/3] RUN rm -f /etc/apt/apt.conf.d/docker-clean;     echo 'Binary::a  0.0s
     => CACHED [build-base-ubuntu 3/3] RUN --mount=type=cache,id=apt-cache-build,target=/var/cache/apt  0.0s
     => CACHED [build-base 1/3] RUN /usr/sbin/update-ccache-symlinks                                    0.0s
     => CACHED [build-base 2/3] RUN mkdir /opt/ccache && ccache --set-config=cache_dir=/opt/ccache &&   0.0s
     => CACHED [build-base 3/3] RUN curl -fsSL -v -o ~/miniconda.sh -O  https://repo.anaconda.com/mini  0.0s
     => CACHED [build-install 1/6] RUN --mount=type=cache,id=conda-build,target=/opt/conda/pkgs     co  0.0s
     => CACHED [build-install 2/6] WORKDIR /opt                                                         0.0s
     => CACHED [build-install 3/6] RUN git clone --recursive --jobs 0 https://github.com/pytorch/pytor  0.0s
     => CACHED [build-install 4/6] RUN git clone --recursive --jobs 0 https://github.com/pytorch/visio  0.0s
     => CACHED [build-install 5/6] RUN git clone --recursive --jobs 0 https://github.com/pytorch/text.  0.0s
     => CACHED [build-install 6/6] RUN git clone --recursive --jobs 0 https://github.com/pytorch/audio  0.0s
     => CACHED [train-builds 2/5] COPY --from=build-install /opt/conda /opt/conda                       0.0s
     => CACHED [build-torch 1/4] WORKDIR /opt/pytorch                                                   0.0s
     => CACHED [build-torch 2/4] RUN if [ -n v1.9.1 ]; then     git checkout v1.9.1 &&     git submodu  0.0s
     => ERROR [build-torch 3/4] RUN --mount=type=cache,target=/opt/ccache     USE_CUDA=1 USE_CUDNN=1   28.3s
    ------                                                                                                   
     > [build-torch 3/4] RUN --mount=type=cache,target=/opt/ccache     USE_CUDA=1 USE_CUDNN=1     TORCH_NVCC_FLAGS=-Xfatbin -compress-all     TORCH_CUDA_ARCH_LIST=8.0     CMAKE_PREFIX_PATH="$(dirname $(which conda))/../"     python setup.py bdist_wheel -d /tmp/dist:                                                     
    #24 0.500 Building wheel torch-1.9.0a0+gitdfbd030                                                        
    #24 0.534 -- Building version 1.9.0a0+gitdfbd030                                                         
    #24 0.549 cmake -GNinja -DBUILD_PYTHON=True -DBUILD_TEST=True -DCMAKE_BUILD_TYPE=Release -DCMAKE_INSTALL_PREFIX=/opt/pytorch/torch -DCMAKE_PREFIX_PATH=/opt/conda/bin/../ -DNUMPY_INCLUDE_DIR=/opt/conda/lib/python3.9/site-packages/numpy/core/include -DPYTHON_EXECUTABLE=/opt/conda/bin/python -DPYTHON_INCLUDE_DIR=/opt/conda/include/python3.9 -DPYTHON_LIBRARY=/opt/conda/lib/libpython3.9.a -DTORCH_BUILD_VERSION=1.9.0a0+gitdfbd030 -DUSE_CUDA=1 -DUSE_CUDNN=1 -DUSE_NUMPY=True /opt/pytorch
    #24 0.647 -- The CXX compiler identification is GNU 7.5.0
    #24 0.716 -- The C compiler identification is GNU 7.5.0
    #24 0.729 -- Detecting CXX compiler ABI info
    #24 0.810 -- Detecting CXX compiler ABI info - done
    #24 0.825 -- Check for working CXX compiler: /usr/bin/c++ - skipped
    #24 0.825 -- Detecting CXX compile features
    #24 0.826 -- Detecting CXX compile features - done
    #24 0.829 -- Detecting C compiler ABI info
    #24 0.900 -- Detecting C compiler ABI info - done
    #24 0.913 -- Check for working C compiler: /usr/bin/cc - skipped
    #24 0.913 -- Detecting C compile features
    #24 0.915 -- Detecting C compile features - done
    #24 0.916 -- Not forcing any particular BLAS to be found
    #24 0.926 -- Performing Test COMPILER_WORKS
    #24 1.001 -- Performing Test COMPILER_WORKS - Success
    #24 1.001 -- Performing Test SUPPORT_GLIBCXX_USE_C99
    #24 1.211 -- Performing Test SUPPORT_GLIBCXX_USE_C99 - Success
    #24 1.212 -- Performing Test CAFFE2_EXCEPTION_PTR_SUPPORTED
    #24 1.409 -- Performing Test CAFFE2_EXCEPTION_PTR_SUPPORTED - Success
    #24 1.410 -- std::exception_ptr is supported.
    #24 1.410 -- Performing Test CAFFE2_NEED_TO_TURN_OFF_DEPRECATION_WARNING
    #24 1.446 -- Performing Test CAFFE2_NEED_TO_TURN_OFF_DEPRECATION_WARNING - Failed
    #24 1.446 -- Turning off deprecation warning due to glog.
    #24 1.446 -- Performing Test CAFFE2_COMPILER_SUPPORTS_AVX2_EXTENSIONS
    #24 1.696 -- Performing Test CAFFE2_COMPILER_SUPPORTS_AVX2_EXTENSIONS - Success
    #24 1.696 -- Current compiler supports avx2 extension. Will build perfkernels.
    #24 1.697 -- Performing Test CAFFE2_COMPILER_SUPPORTS_AVX512_EXTENSIONS
    #24 1.904 -- Performing Test CAFFE2_COMPILER_SUPPORTS_AVX512_EXTENSIONS - Success
    #24 1.904 -- Current compiler supports avx512f extension. Will build fbgemm.
    #24 1.905 -- Performing Test COMPILER_SUPPORTS_HIDDEN_VISIBILITY
    #24 1.988 -- Performing Test COMPILER_SUPPORTS_HIDDEN_VISIBILITY - Success
    #24 1.990 -- Performing Test COMPILER_SUPPORTS_HIDDEN_INLINE_VISIBILITY
    #24 2.068 -- Performing Test COMPILER_SUPPORTS_HIDDEN_INLINE_VISIBILITY - Success
    #24 2.070 -- Performing Test COMPILER_SUPPORTS_RDYNAMIC
    #24 2.152 -- Performing Test COMPILER_SUPPORTS_RDYNAMIC - Success
    #24 2.170 -- Building using own protobuf under third_party per request.
    #24 2.170 -- Use custom protobuf build.
    #24 2.171 -- 
    #24 2.171 -- 3.11.4.0
    #24 2.172 -- Looking for pthread.h
    #24 2.248 -- Looking for pthread.h - found
    #24 2.248 -- Performing Test CMAKE_HAVE_LIBC_PTHREAD
    #24 2.327 -- Performing Test CMAKE_HAVE_LIBC_PTHREAD - Failed
    #24 2.327 -- Check if compiler accepts -pthread
    #24 2.402 -- Check if compiler accepts -pthread - yes
    #24 2.404 -- Found Threads: TRUE  
    #24 2.405 -- Performing Test protobuf_HAVE_BUILTIN_ATOMICS
    #24 2.512 -- Performing Test protobuf_HAVE_BUILTIN_ATOMICS - Success
    #24 2.529 -- Caffe2 protobuf include directory: $<BUILD_INTERFACE:/opt/pytorch/third_party/protobuf/src>$<INSTALL_INTERFACE:include>
    #24 2.530 -- Trying to find preferred BLAS backend of choice: MKL
    #24 2.531 -- MKL_THREADING = OMP
    #24 2.532 -- Looking for sys/types.h
    #24 2.604 -- Looking for sys/types.h - found
    #24 2.605 -- Looking for stdint.h
    #24 2.672 -- Looking for stdint.h - found
    #24 2.673 -- Looking for stddef.h
    #24 2.743 -- Looking for stddef.h - found
    #24 2.743 -- Check size of void*
    #24 2.816 -- Check size of void* - done
    #24 3.083 CMake Warning (dev) at /opt/conda/share/cmake-3.19/Modules/FindPackageHandleStandardArgs.cmake:426 (message):
    #24 3.083   The package name passed to `find_package_handle_standard_args` (OpenMP_C)
    #24 3.083   does not match the name of the calling package (OpenMP).  This can lead to
    #24 3.083   problems in calling code that expects `find_package` result variables
    #24 3.083   (e.g., `_FOUND`) to follow a certain pattern.
    #24 3.083 Call Stack (most recent call first):
    #24 3.083   cmake/Modules/FindOpenMP.cmake:576 (find_package_handle_standard_args)
    #24 3.083   cmake/Modules/FindMKL.cmake:213 (FIND_PACKAGE)
    #24 3.083   cmake/Modules/FindMKL.cmake:307 (CHECK_ALL_LIBRARIES)
    #24 3.083   cmake/Dependencies.cmake:144 (find_package)
    #24 3.083   CMakeLists.txt:621 (include)
    #24 3.083 This warning is for project developers.  Use -Wno-dev to suppress it.
    #24 3.083 
    #24 3.167 CMake Warning (dev) at /opt/conda/share/cmake-3.19/Modules/FindPackageHandleStandardArgs.cmake:426 (message):
    #24 3.167   The package name passed to `find_package_handle_standard_args` (OpenMP_CXX)
    #24 3.167   does not match the name of the calling package (OpenMP).  This can lead to
    #24 3.167   problems in calling code that expects `find_package` result variables
    #24 3.167   (e.g., `_FOUND`) to follow a certain pattern.
    #24 3.167 Call Stack (most recent call first):
    #24 3.167   cmake/Modules/FindOpenMP.cmake:576 (find_package_handle_standard_args)
    #24 3.167   cmake/Modules/FindMKL.cmake:213 (FIND_PACKAGE)
    #24 3.167   cmake/Modules/FindMKL.cmake:307 (CHECK_ALL_LIBRARIES)
    #24 3.167   cmake/Dependencies.cmake:144 (find_package)
    #24 3.167   CMakeLists.txt:621 (include)
    #24 3.167 This warning is for project developers.  Use -Wno-dev to suppress it.
    #24 3.167 
    #24 3.172 -- Looking for cblas_sgemm
    #24 3.420 -- Looking for cblas_sgemm - found
    #24 3.426 -- MKL libraries: /opt/conda/lib/libmkl_intel_lp64.so;/opt/conda/lib/libmkl_gnu_thread.so;/opt/conda/lib/libmkl_core.so;-fopenmp;/usr/lib/x86_64-linux-gnu/libpthread.so;/usr/lib/x86_64-linux-gnu/libm.so;/usr/lib/x86_64-linux-gnu/libdl.so
    #24 3.426 -- MKL include directory: /opt/conda/include
    #24 3.426 -- MKL OpenMP type: GNU
    #24 3.426 -- MKL OpenMP library: -fopenmp
    #24 3.449 -- The ASM compiler identification is GNU
    #24 3.452 -- Found assembler: /usr/bin/cc
    #24 3.467 -- Brace yourself, we are building NNPACK
    #24 3.471 -- Performing Test NNPACK_ARCH_IS_X86_32
    #24 3.506 -- Performing Test NNPACK_ARCH_IS_X86_32 - Failed
    #24 3.529 -- Found PythonInterp: /opt/conda/bin/python (found version "3.9.7") 
    #24 3.529 -- NNPACK backend is x86-64
    #24 3.553 CMake Deprecation Warning at third_party/googletest/CMakeLists.txt:1 (cmake_minimum_required):
    #24 3.553   Compatibility with CMake < 2.8.12 will be removed from a future version of
    #24 3.553   CMake.
    #24 3.553 
    #24 3.553   Update the VERSION argument <min> value or use a ...<max> suffix to tell
    #24 3.553   CMake that the project does not need compatibility with older versions.
    #24 3.553 
    #24 3.553 
    #24 3.555 CMake Deprecation Warning at third_party/googletest/googlemock/CMakeLists.txt:42 (cmake_minimum_required):
    #24 3.555   Compatibility with CMake < 2.8.12 will be removed from a future version of
    #24 3.555   CMake.
    #24 3.555 
    #24 3.555   Update the VERSION argument <min> value or use a ...<max> suffix to tell
    #24 3.555   CMake that the project does not need compatibility with older versions.
    #24 3.555 
    #24 3.555 
    #24 3.556 CMake Deprecation Warning at third_party/googletest/googletest/CMakeLists.txt:49 (cmake_minimum_required):
    #24 3.556   Compatibility with CMake < 2.8.12 will be removed from a future version of
    #24 3.556   CMake.
    #24 3.556 
    #24 3.556   Update the VERSION argument <min> value or use a ...<max> suffix to tell
    #24 3.556   CMake that the project does not need compatibility with older versions.
    #24 3.556 
    #24 3.556 
    #24 3.586 -- Failed to find LLVM FileCheck
    #24 3.592 -- Found Git: /usr/bin/git (found version "2.17.1") 
    #24 3.614 -- git Version: v1.4.0-505be96a
    #24 3.614 -- Version: 1.4.0
    #24 3.617 -- Performing Test HAVE_CXX_FLAG_STD_CXX11
    #24 3.700 -- Performing Test HAVE_CXX_FLAG_STD_CXX11 - Success
    #24 3.702 -- Performing Test HAVE_CXX_FLAG_WALL
    #24 3.783 -- Performing Test HAVE_CXX_FLAG_WALL - Success
    #24 3.785 -- Performing Test HAVE_CXX_FLAG_WEXTRA
    #24 3.868 -- Performing Test HAVE_CXX_FLAG_WEXTRA - Success
    #24 3.870 -- Performing Test HAVE_CXX_FLAG_WSHADOW
    #24 3.952 -- Performing Test HAVE_CXX_FLAG_WSHADOW - Success
    #24 3.954 -- Performing Test HAVE_CXX_FLAG_WERROR
    #24 4.031 -- Performing Test HAVE_CXX_FLAG_WERROR - Success
    #24 4.033 -- Performing Test HAVE_CXX_FLAG_PEDANTIC
    #24 4.112 -- Performing Test HAVE_CXX_FLAG_PEDANTIC - Success
    #24 4.114 -- Performing Test HAVE_CXX_FLAG_PEDANTIC_ERRORS
    #24 4.197 -- Performing Test HAVE_CXX_FLAG_PEDANTIC_ERRORS - Success
    #24 4.199 -- Performing Test HAVE_CXX_FLAG_WSHORTEN_64_TO_32
    #24 4.240 -- Performing Test HAVE_CXX_FLAG_WSHORTEN_64_TO_32 - Failed
    #24 4.240 -- Performing Test HAVE_CXX_FLAG_WFLOAT_EQUAL
    #24 4.320 -- Performing Test HAVE_CXX_FLAG_WFLOAT_EQUAL - Success
    #24 4.322 -- Performing Test HAVE_CXX_FLAG_FSTRICT_ALIASING
    #24 4.402 -- Performing Test HAVE_CXX_FLAG_FSTRICT_ALIASING - Success
    #24 4.404 -- Performing Test HAVE_CXX_FLAG_WNO_DEPRECATED_DECLARATIONS
    #24 4.484 -- Performing Test HAVE_CXX_FLAG_WNO_DEPRECATED_DECLARATIONS - Success
    #24 4.486 -- Performing Test HAVE_CXX_FLAG_WSTRICT_ALIASING
    #24 4.566 -- Performing Test HAVE_CXX_FLAG_WSTRICT_ALIASING - Success
    #24 4.568 -- Performing Test HAVE_CXX_FLAG_WD654
    #24 4.598 -- Performing Test HAVE_CXX_FLAG_WD654 - Failed
    #24 4.600 -- Performing Test HAVE_CXX_FLAG_WTHREAD_SAFETY
    #24 4.634 -- Performing Test HAVE_CXX_FLAG_WTHREAD_SAFETY - Failed
    #24 4.635 -- Performing Test HAVE_CXX_FLAG_COVERAGE
    #24 4.720 -- Performing Test HAVE_CXX_FLAG_COVERAGE - Success
    #24 4.720 -- Performing Test HAVE_STD_REGEX
    #24 4.720 -- Performing Test HAVE_STD_REGEX
    #24 6.312 -- Performing Test HAVE_STD_REGEX -- success
    #24 6.312 -- Performing Test HAVE_GNU_POSIX_REGEX
    #24 6.312 -- Performing Test HAVE_GNU_POSIX_REGEX
    #24 6.350 -- Performing Test HAVE_GNU_POSIX_REGEX -- failed to compile
    #24 6.350 -- Performing Test HAVE_POSIX_REGEX
    #24 6.350 -- Performing Test HAVE_POSIX_REGEX
    #24 6.543 -- Performing Test HAVE_POSIX_REGEX -- success
    #24 6.543 -- Performing Test HAVE_STEADY_CLOCK
    #24 6.543 -- Performing Test HAVE_STEADY_CLOCK
    #24 6.681 -- Performing Test HAVE_STEADY_CLOCK -- success
    #24 6.714 -- Performing Test COMPILER_SUPPORTS_AVX512
    #24 6.798 -- Performing Test COMPILER_SUPPORTS_AVX512 - Success
    #24 6.802 CMake Warning (dev) at /opt/conda/share/cmake-3.19/Modules/FindPackageHandleStandardArgs.cmake:426 (message):
    #24 6.802   The package name passed to `find_package_handle_standard_args` (OpenMP_C)
    #24 6.802   does not match the name of the calling package (OpenMP).  This can lead to
    #24 6.802   problems in calling code that expects `find_package` result variables
    #24 6.802   (e.g., `_FOUND`) to follow a certain pattern.
    #24 6.802 Call Stack (most recent call first):
    #24 6.802   cmake/Modules/FindOpenMP.cmake:576 (find_package_handle_standard_args)
    #24 6.802   third_party/fbgemm/CMakeLists.txt:60 (find_package)
    #24 6.802 This warning is for project developers.  Use -Wno-dev to suppress it.
    #24 6.802 
    #24 6.803 -- Found OpenMP_C: -fopenmp (found version "4.5") 
    #24 6.803 CMake Warning (dev) at /opt/conda/share/cmake-3.19/Modules/FindPackageHandleStandardArgs.cmake:426 (message):
    #24 6.803   The package name passed to `find_package_handle_standard_args` (OpenMP_CXX)
    #24 6.803   does not match the name of the calling package (OpenMP).  This can lead to
    #24 6.803   problems in calling code that expects `find_package` result variables
    #24 6.803   (e.g., `_FOUND`) to follow a certain pattern.
    #24 6.803 Call Stack (most recent call first):
    #24 6.803   cmake/Modules/FindOpenMP.cmake:576 (find_package_handle_standard_args)
    #24 6.803   third_party/fbgemm/CMakeLists.txt:60 (find_package)
    #24 6.803 This warning is for project developers.  Use -Wno-dev to suppress it.
    #24 6.803 
    #24 6.804 -- Found OpenMP_CXX: -fopenmp (found version "4.5") 
    #24 6.804 -- Found OpenMP: TRUE (found version "4.5")  
    #24 6.804 CMake Warning at third_party/fbgemm/CMakeLists.txt:62 (message):
    #24 6.804   OpenMP found! OpenMP_C_INCLUDE_DIRS =
    #24 6.804 
    #24 6.804 
    #24 6.916 CMake Warning at third_party/fbgemm/CMakeLists.txt:142 (message):
    #24 6.916   ==========
    #24 6.916 
    #24 6.916 
    #24 6.916 CMake Warning at third_party/fbgemm/CMakeLists.txt:143 (message):
    #24 6.916   CMAKE_BUILD_TYPE = Release
    #24 6.916 
    #24 6.916 
    #24 6.916 CMake Warning at third_party/fbgemm/CMakeLists.txt:144 (message):
    #24 6.916   CMAKE_CXX_FLAGS_DEBUG is -g
    #24 6.916 
    #24 6.916 
    #24 6.916 CMake Warning at third_party/fbgemm/CMakeLists.txt:145 (message):
    #24 6.916   CMAKE_CXX_FLAGS_RELEASE is -O3 -DNDEBUG
    #24 6.916 
    #24 6.916 
    #24 6.916 CMake Warning at third_party/fbgemm/CMakeLists.txt:146 (message):
    #24 6.916   ==========
    #24 6.916 
    #24 6.916 
    #24 6.923 -- Performing Test __CxxFlag__fno_threadsafe_statics
    #24 7.006 -- Performing Test __CxxFlag__fno_threadsafe_statics - Success
    #24 7.008 -- Performing Test __CxxFlag__fno_semantic_interposition
    #24 7.087 -- Performing Test __CxxFlag__fno_semantic_interposition - Success
    #24 7.088 -- Performing Test __CxxFlag__fmerge_all_constants
    #24 7.172 -- Performing Test __CxxFlag__fmerge_all_constants - Success
    #24 7.174 -- Performing Test __CxxFlag__fno_enforce_eh_specs
    #24 7.254 -- Performing Test __CxxFlag__fno_enforce_eh_specs - Success
    #24 7.258 ** AsmJit Summary **
    #24 7.258    ASMJIT_DIR=/opt/pytorch/third_party/fbgemm/third_party/asmjit
    #24 7.258    ASMJIT_TEST=FALSE
    #24 7.258    ASMJIT_TARGET_TYPE=STATIC
    #24 7.258    ASMJIT_DEPS=pthread;rt
    #24 7.258    ASMJIT_LIBS=asmjit;pthread;rt
    #24 7.258    ASMJIT_CFLAGS=-DASMJIT_STATIC
    #24 7.258    ASMJIT_PRIVATE_CFLAGS=-Wall;-Wextra;-Wconversion;-fno-math-errno;-fno-threadsafe-statics;-fno-semantic-interposition;-DASMJIT_STATIC
    #24 7.258    ASMJIT_PRIVATE_CFLAGS_DBG=
    #24 7.258    ASMJIT_PRIVATE_CFLAGS_REL=-O2;-fmerge-all-constants;-fno-enforce-eh-specs
    #24 7.267 CMake Warning at cmake/Dependencies.cmake:793 (message):
    #24 7.267   Not compiling with NUMA.  Suppress this warning with -DUSE_NUMA=OFF
    #24 7.267 Call Stack (most recent call first):
    #24 7.267   CMakeLists.txt:621 (include)
    #24 7.267 
    #24 7.267 
    #24 7.267 -- Could NOT find Numa (missing: Numa_INCLUDE_DIR Numa_LIBRARIES) 
    #24 7.268 -- Using third party subdirectory Eigen.
    #24 7.291 -- Found PythonInterp: /opt/conda/bin/python (found suitable version "3.9.7", minimum required is "3.0") 
    #24 7.296 -- Found PythonLibs: /opt/conda/lib/libpython3.9.a (found suitable version "3.9.7", minimum required is "3.0") 
    #24 7.301 -- Could NOT find pybind11 (missing: pybind11_DIR)
    #24 7.302 -- Could NOT find pybind11 (missing: pybind11_INCLUDE_DIR) 
    #24 7.302 -- Using third_party/pybind11.
    #24 7.303 -- pybind11 include dirs: /opt/pytorch/cmake/../third_party/pybind11/include
    #24 7.426 -- Could NOT find MPI_C (missing: MPI_C_LIB_NAMES MPI_C_HEADER_DIR MPI_C_WORKS) 
    #24 7.506 -- Could NOT find MPI_CXX (missing: MPI_CXX_LIB_NAMES MPI_CXX_HEADER_DIR MPI_CXX_WORKS) 
    #24 7.507 -- Could NOT find MPI (missing: MPI_C_FOUND MPI_CXX_FOUND) 
    #24 7.507     Reason given by package: MPI component 'Fortran' was requested, but language Fortran is not enabled.  
    #24 7.507 
    #24 7.507 CMake Warning at cmake/Dependencies.cmake:1050 (message):
    #24 7.507   Not compiling with MPI.  Suppress this warning with -DUSE_MPI=OFF
    #24 7.507 Call Stack (most recent call first):
    #24 7.507   CMakeLists.txt:621 (include)
    #24 7.507 
    #24 7.507 
    #24 7.509 CMake Warning (dev) at /opt/conda/share/cmake-3.19/Modules/FindPackageHandleStandardArgs.cmake:426 (message):
    #24 7.509   The package name passed to `find_package_handle_standard_args` (OpenMP_C)
    #24 7.509   does not match the name of the calling package (OpenMP).  This can lead to
    #24 7.509   problems in calling code that expects `find_package` result variables
    #24 7.509   (e.g., `_FOUND`) to follow a certain pattern.
    #24 7.509 Call Stack (most recent call first):
    #24 7.509   cmake/Modules/FindOpenMP.cmake:576 (find_package_handle_standard_args)
    #24 7.509   cmake/Dependencies.cmake:1105 (find_package)
    #24 7.509   CMakeLists.txt:621 (include)
    #24 7.509 This warning is for project developers.  Use -Wno-dev to suppress it.
    #24 7.509 
    #24 7.509 CMake Warning (dev) at /opt/conda/share/cmake-3.19/Modules/FindPackageHandleStandardArgs.cmake:426 (message):
    #24 7.509   The package name passed to `find_package_handle_standard_args` (OpenMP_CXX)
    #24 7.509   does not match the name of the calling package (OpenMP).  This can lead to
    #24 7.509   problems in calling code that expects `find_package` result variables
    #24 7.509   (e.g., `_FOUND`) to follow a certain pattern.
    #24 7.509 Call Stack (most recent call first):
    #24 7.509   cmake/Modules/FindOpenMP.cmake:576 (find_package_handle_standard_args)
    #24 7.509   cmake/Dependencies.cmake:1105 (find_package)
    #24 7.509   CMakeLists.txt:621 (include)
    #24 7.509 This warning is for project developers.  Use -Wno-dev to suppress it.
    #24 7.509 
    #24 7.510 -- Adding OpenMP CXX_FLAGS: -fopenmp
    #24 7.510 -- Will link against OpenMP libraries: /usr/lib/gcc/x86_64-linux-gnu/7/libgomp.so;/usr/lib/x86_64-linux-gnu/libpthread.so
    #24 7.523 -- Found CUDA: /usr/local/cuda (found version "10.2") 
    #24 7.525 -- Caffe2: CUDA detected: 10.2
    #24 7.525 -- Caffe2: CUDA nvcc is: /usr/local/cuda/bin/nvcc
    #24 7.525 -- Caffe2: CUDA toolkit directory: /usr/local/cuda
    #24 7.628 -- Caffe2: Header version is: 10.2
    #24 7.633 -- Found CUDNN: /usr/lib/x86_64-linux-gnu/libcudnn.so  
    #24 7.633 -- Found cuDNN: v8.2.0  (include: /usr/include, library: /usr/lib/x86_64-linux-gnu/libcudnn.so)
    #24 7.677 -- /usr/local/cuda/lib64/libnvrtc.so shorthash is 08c4863f
    #24 7.678 CMake Warning at cmake/public/utils.cmake:365 (message):
    #24 7.678   In the future we will require one to explicitly pass TORCH_CUDA_ARCH_LIST
    #24 7.678   to cmake instead of implicitly setting it as an env variable.  This will
    #24 7.678   become a FATAL_ERROR in future version of pytorch.
    #24 7.678 Call Stack (most recent call first):
    #24 7.678   cmake/public/cuda.cmake:511 (torch_cuda_get_nvcc_gencode_flag)
    #24 7.678   cmake/Dependencies.cmake:1155 (include)
    #24 7.678   CMakeLists.txt:621 (include)
    #24 7.678 
    #24 7.678 
    #24 7.679 -- Added CUDA NVCC flags for: -gencode;arch=compute_80,code=sm_80
    #24 7.680 CMake Warning at cmake/public/utils.cmake:365 (message):
    #24 7.680   In the future we will require one to explicitly pass TORCH_CUDA_ARCH_LIST
    #24 7.680   to cmake instead of implicitly setting it as an env variable.  This will
    #24 7.680   become a FATAL_ERROR in future version of pytorch.
    #24 7.680 Call Stack (most recent call first):
    #24 7.680   cmake/External/nccl.cmake:13 (torch_cuda_get_nvcc_gencode_flag)
    #24 7.680   cmake/Dependencies.cmake:1288 (include)
    #24 7.680   CMakeLists.txt:621 (include)
    #24 7.680 
    #24 7.680 
    #24 7.698 CMake Warning at cmake/External/nccl.cmake:62 (message):
    #24 7.698   Objcopy version is too old to support NCCL library slimming
    #24 7.698 Call Stack (most recent call first):
    #24 7.698   cmake/Dependencies.cmake:1288 (include)
    #24 7.698   CMakeLists.txt:621 (include)
    #24 7.698 
    #24 7.698 
    #24 7.700 -- Could NOT find CUB (missing: CUB_INCLUDE_DIR) 
    #24 7.703 CMake Warning (dev) at third_party/gloo/CMakeLists.txt:21 (option):
    #24 7.703   Policy CMP0077 is not set: option() honors normal variables.  Run "cmake
    #24 7.703   --help-policy CMP0077" for policy details.  Use the cmake_policy command to
    #24 7.703   set the policy and suppress this warning.
    #24 7.703 
    #24 7.703   For compatibility with older versions of CMake, option is clearing the
    #24 7.703   normal variable 'BUILD_BENCHMARK'.
    #24 7.703 This warning is for project developers.  Use -Wno-dev to suppress it.
    #24 7.703 
    #24 7.703 -- Gloo build as SHARED library
    #24 7.711 -- Found CUDA: /usr/local/cuda (found suitable version "10.2", minimum required is "7.0") 
    #24 7.712 -- CUDA detected: 10.2
    #24 7.714 CMake Warning (dev) at /opt/conda/share/cmake-3.19/Modules/FindPackageHandleStandardArgs.cmake:426 (message):
    #24 7.714   The package name passed to `find_package_handle_standard_args` (NCCL) does
    #24 7.714   not match the name of the calling package (nccl).  This can lead to
    #24 7.714   problems in calling code that expects `find_package` result variables
    #24 7.714   (e.g., `_FOUND`) to follow a certain pattern.
    #24 7.714 Call Stack (most recent call first):
    #24 7.714   third_party/gloo/cmake/Modules/Findnccl.cmake:45 (find_package_handle_standard_args)
    #24 7.714   third_party/gloo/cmake/Dependencies.cmake:128 (find_package)
    #24 7.714   third_party/gloo/CMakeLists.txt:109 (include)
    #24 7.714 This warning is for project developers.  Use -Wno-dev to suppress it.
    #24 7.714 
    #24 7.714 -- Found NCCL: /usr/include  
    #24 7.714 -- Determining NCCL version from the header file: /usr/include/nccl.h
    #24 7.714 -- NCCL_MAJOR_VERSION: 2
    #24 7.714 -- Found NCCL (include: /usr/include, library: /usr/lib/x86_64-linux-gnu/libnccl.so)
    #24 7.733 -- Found CUDA: /usr/local/cuda (found version "10.2") 
    #24 7.779 -- Performing Test UV_LINT_W4
    #24 7.806 -- Performing Test UV_LINT_W4 - Failed
    #24 7.808 -- Performing Test UV_LINT_NO_UNUSED_PARAMETER_MSVC
    #24 7.836 -- Performing Test UV_LINT_NO_UNUSED_PARAMETER_MSVC - Failed
    #24 7.837 -- Performing Test UV_LINT_NO_CONDITIONAL_CONSTANT_MSVC
    #24 7.866 -- Performing Test UV_LINT_NO_CONDITIONAL_CONSTANT_MSVC - Failed
    #24 7.867 -- Performing Test UV_LINT_NO_NONSTANDARD_MSVC
    #24 7.895 -- Performing Test UV_LINT_NO_NONSTANDARD_MSVC - Failed
    #24 7.896 -- Performing Test UV_LINT_NO_NONSTANDARD_EMPTY_TU_MSVC
    #24 7.923 -- Performing Test UV_LINT_NO_NONSTANDARD_EMPTY_TU_MSVC - Failed
    #24 7.924 -- Performing Test UV_LINT_NO_NONSTANDARD_FILE_SCOPE_MSVC
    #24 7.951 -- Performing Test UV_LINT_NO_NONSTANDARD_FILE_SCOPE_MSVC - Failed
    #24 7.953 -- Performing Test UV_LINT_NO_NONSTANDARD_NONSTATIC_DLIMPORT_MSVC
    #24 7.979 -- Performing Test UV_LINT_NO_NONSTANDARD_NONSTATIC_DLIMPORT_MSVC - Failed
    #24 7.981 -- Performing Test UV_LINT_NO_HIDES_LOCAL
    #24 8.006 -- Performing Test UV_LINT_NO_HIDES_LOCAL - Failed
    #24 8.008 -- Performing Test UV_LINT_NO_HIDES_PARAM
    #24 8.034 -- Performing Test UV_LINT_NO_HIDES_PARAM - Failed
    #24 8.036 -- Performing Test UV_LINT_NO_HIDES_GLOBAL
    #24 8.061 -- Performing Test UV_LINT_NO_HIDES_GLOBAL - Failed
    #24 8.063 -- Performing Test UV_LINT_NO_CONDITIONAL_ASSIGNMENT_MSVC
    #24 8.091 -- Performing Test UV_LINT_NO_CONDITIONAL_ASSIGNMENT_MSVC - Failed
    #24 8.092 -- Performing Test UV_LINT_NO_UNSAFE_MSVC
    #24 8.117 -- Performing Test UV_LINT_NO_UNSAFE_MSVC - Failed
    #24 8.118 -- Performing Test UV_LINT_WALL
    #24 8.190 -- Performing Test UV_LINT_WALL - Success
    #24 8.192 -- Performing Test UV_LINT_NO_UNUSED_PARAMETER
    #24 8.263 -- Performing Test UV_LINT_NO_UNUSED_PARAMETER - Success
    #24 8.265 -- Performing Test UV_LINT_STRICT_PROTOTYPES
    #24 8.340 -- Performing Test UV_LINT_STRICT_PROTOTYPES - Success
    #24 8.341 -- Performing Test UV_LINT_EXTRA
    #24 8.414 -- Performing Test UV_LINT_EXTRA - Success
    #24 8.416 -- Performing Test UV_LINT_UTF8_MSVC
    #24 8.440 -- Performing Test UV_LINT_UTF8_MSVC - Failed
    #24 8.442 -- Performing Test UV_F_STRICT_ALIASING
    #24 8.512 -- Performing Test UV_F_STRICT_ALIASING - Success
    #24 8.514 -- summary of build options:
    #24 8.514     Install prefix:  /opt/pytorch/torch
    #24 8.514     Target system:   Linux
    #24 8.514     Compiler:
    #24 8.514       C compiler:    /usr/bin/cc
    #24 8.514       CFLAGS:          -fopenmp
    #24 8.514 
    #24 8.517 -- Found uv: 1.38.1 (found version "1.38.1") 
    #24 8.519 CMake Warning at cmake/Dependencies.cmake:1406 (message):
    #24 8.519   Metal is only used in ios builds.
    #24 8.519 Call Stack (most recent call first):
    #24 8.519   CMakeLists.txt:621 (include)
    #24 8.519 
    #24 8.519 
    #24 8.522 Generated: /opt/pytorch/build/third_party/onnx/onnx/onnx_onnx_torch-ml.proto
    #24 8.522 Generated: /opt/pytorch/build/third_party/onnx/onnx/onnx-operators_onnx_torch-ml.proto
    #24 8.523 Generated: /opt/pytorch/build/third_party/onnx/onnx/onnx-data_onnx_torch.proto
    #24 8.626 -- 
    #24 8.626 -- ******** Summary ********
    #24 8.626 --   CMake version         : 3.19.6
    #24 8.626 --   CMake command         : /opt/conda/bin/cmake
    #24 8.626 --   System                : Linux
    #24 8.626 --   C++ compiler          : /usr/bin/c++
    #24 8.626 --   C++ compiler version  : 7.5.0
    #24 8.626 --   CXX flags             :  -Wno-deprecated -fvisibility-inlines-hidden -DUSE_PTHREADPOOL -fopenmp -Wnon-virtual-dtor
    #24 8.626 --   Build type            : Release
    #24 8.626 --   Compile definitions   : TH_BLAS_MKL;ONNX_ML=1;ONNXIFI_ENABLE_EXT=1
    #24 8.626 --   CMAKE_PREFIX_PATH     : /opt/conda/bin/../;/usr/local/cuda
    #24 8.626 --   CMAKE_INSTALL_PREFIX  : /opt/pytorch/torch
    #24 8.626 --   CMAKE_MODULE_PATH     : /opt/pytorch/cmake/Modules;/opt/pytorch/cmake/public/../Modules_CUDA_fix
    #24 8.626 -- 
    #24 8.626 --   ONNX version          : 1.8.0
    #24 8.626 --   ONNX NAMESPACE        : onnx_torch
    #24 8.626 --   ONNX_BUILD_TESTS      : OFF
    #24 8.626 --   ONNX_BUILD_BENCHMARKS : OFF
    #24 8.626 --   ONNX_USE_LITE_PROTO   : OFF
    #24 8.626 --   ONNXIFI_DUMMY_BACKEND : OFF
    #24 8.626 --   ONNXIFI_ENABLE_EXT    : OFF
    #24 8.626 -- 
    #24 8.626 --   Protobuf compiler     : 
    #24 8.626 --   Protobuf includes     : 
    #24 8.626 --   Protobuf libraries    : 
    #24 8.626 --   BUILD_ONNX_PYTHON     : OFF
    #24 8.627 -- 
    #24 8.627 -- ******** Summary ********
    #24 8.627 --   CMake version         : 3.19.6
    #24 8.627 --   CMake command         : /opt/conda/bin/cmake
    #24 8.627 --   System                : Linux
    #24 8.627 --   C++ compiler          : /usr/bin/c++
    #24 8.627 --   C++ compiler version  : 7.5.0
    #24 8.627 --   CXX flags             :  -Wno-deprecated -fvisibility-inlines-hidden -DUSE_PTHREADPOOL -fopenmp -Wnon-virtual-dtor
    #24 8.627 --   Build type            : Release
    #24 8.627 --   Compile definitions   : TH_BLAS_MKL;ONNX_ML=1;ONNXIFI_ENABLE_EXT=1
    #24 8.627 --   CMAKE_PREFIX_PATH     : /opt/conda/bin/../;/usr/local/cuda
    #24 8.627 --   CMAKE_INSTALL_PREFIX  : /opt/pytorch/torch
    #24 8.627 --   CMAKE_MODULE_PATH     : /opt/pytorch/cmake/Modules;/opt/pytorch/cmake/public/../Modules_CUDA_fix
    #24 8.627 -- 
    #24 8.627 --   ONNX version          : 1.4.1
    #24 8.627 --   ONNX NAMESPACE        : onnx_torch
    #24 8.627 --   ONNX_BUILD_TESTS      : OFF
    #24 8.627 --   ONNX_BUILD_BENCHMARKS : OFF
    #24 8.627 --   ONNX_USE_LITE_PROTO   : OFF
    #24 8.627 --   ONNXIFI_DUMMY_BACKEND : OFF
    #24 8.627 -- 
    #24 8.627 --   Protobuf compiler     : 
    #24 8.627 --   Protobuf includes     : 
    #24 8.627 --   Protobuf libraries    : 
    #24 8.627 --   BUILD_ONNX_PYTHON     : OFF
    #24 8.628 -- Found CUDA with FP16 support, compiling with torch.cuda.HalfTensor
    #24 8.628 -- Adding -DNDEBUG to compile flags
    #24 8.629 -- Checking prototype magma_get_sgeqrf_nb for MAGMA_V2
    #24 8.755 -- Checking prototype magma_get_sgeqrf_nb for MAGMA_V2 - True
    #24 8.755 -- Compiling with MAGMA support
    #24 8.755 -- MAGMA INCLUDE DIRECTORIES: /opt/conda/include
    #24 8.755 -- MAGMA LIBRARIES: /opt/conda/lib/libmagma.a
    #24 8.755 -- MAGMA V2 check: 1
    #24 8.792 -- Could not find hardware support for NEON on this machine.
    #24 8.792 -- No OMAP3 processor on this machine.
    #24 8.792 -- No OMAP4 processor on this machine.
    #24 8.793 -- Looking for cpuid.h
    #24 8.867 -- Looking for cpuid.h - found
    #24 8.868 -- Performing Test HAVE_GCC_GET_CPUID
    #24 8.948 -- Performing Test HAVE_GCC_GET_CPUID - Success
    #24 8.948 -- Performing Test NO_GCC_EBX_FPIC_BUG
    #24 9.022 -- Performing Test NO_GCC_EBX_FPIC_BUG - Success
    #24 9.022 -- <FindVSX>
    #24 9.025 -- Performing Test C_VSX_FOUND
    #24 9.056 -- Performing Test C_VSX_FOUND - Failed
    #24 9.056 -- Performing Test CXX_VSX_FOUND
    #24 9.086 -- Performing Test CXX_VSX_FOUND - Failed
    #24 9.086 -- </FindVSX>
    #24 9.087 -- Performing Test C_HAS_AVX_1
    #24 9.215 -- Performing Test C_HAS_AVX_1 - Failed
    #24 9.215 -- Performing Test C_HAS_AVX_2
    #24 9.380 -- Performing Test C_HAS_AVX_2 - Success
    #24 9.381 -- Performing Test C_HAS_AVX2_1
    #24 9.507 -- Performing Test C_HAS_AVX2_1 - Failed
    #24 9.507 -- Performing Test C_HAS_AVX2_2
    #24 9.665 -- Performing Test C_HAS_AVX2_2 - Success
    #24 9.666 -- Performing Test CXX_HAS_AVX_1
    #24 9.797 -- Performing Test CXX_HAS_AVX_1 - Failed
    #24 9.797 -- Performing Test CXX_HAS_AVX_2
    #24 9.960 -- Performing Test CXX_HAS_AVX_2 - Success
    #24 9.961 -- Performing Test CXX_HAS_AVX2_1
    #24 10.09 -- Performing Test CXX_HAS_AVX2_1 - Failed
    #24 10.09 -- Performing Test CXX_HAS_AVX2_2
    #24 10.25 -- Performing Test CXX_HAS_AVX2_2 - Success
    #24 10.25 -- AVX compiler support found
    #24 10.25 -- AVX2 compiler support found
    #24 10.25 -- Performing Test BLAS_F2C_DOUBLE_WORKS
    #24 10.50 -- Performing Test BLAS_F2C_DOUBLE_WORKS - Failed
    #24 10.50 -- Performing Test BLAS_F2C_FLOAT_WORKS
    #24 10.74 -- Performing Test BLAS_F2C_FLOAT_WORKS - Success
    #24 10.74 -- Performing Test BLAS_USE_CBLAS_DOT
    #24 11.03 -- Performing Test BLAS_USE_CBLAS_DOT - Success
    #24 11.03 -- Found a library with BLAS API (mkl). Full path: (/opt/conda/lib/libmkl_intel_lp64.so;/opt/conda/lib/libmkl_gnu_thread.so;/opt/conda/lib/libmkl_core.so;-fopenmp;/usr/lib/x86_64-linux-gnu/libpthread.so;/usr/lib/x86_64-linux-gnu/libm.so;/usr/lib/x86_64-linux-gnu/libdl.so)
    #24 11.03 -- Found a library with LAPACK API (mkl).
    #24 11.03 disabling ROCM because NOT USE_ROCM is set
    #24 11.03 -- MIOpen not found. Compiling without MIOpen support
    #24 11.04 -- MKLDNN_CPU_RUNTIME = OMP
    #24 11.04 CMake Deprecation Warning at third_party/ideep/mkl-dnn/CMakeLists.txt:17 (cmake_minimum_required):
    #24 11.04   Compatibility with CMake < 2.8.12 will be removed from a future version of
    #24 11.04   CMake.
    #24 11.04 
    #24 11.04   Update the VERSION argument <min> value or use a ...<max> suffix to tell
    #24 11.04   CMake that the project does not need compatibility with older versions.
    #24 11.04 
    #24 11.04 
    #24 11.04 -- Intel MKL-DNN compat: set DNNL_ENABLE_CONCURRENT_EXEC to MKLDNN_ENABLE_CONCURRENT_EXEC with value `ON`
    #24 11.04 -- Intel MKL-DNN compat: set DNNL_BUILD_EXAMPLES to MKLDNN_BUILD_EXAMPLES with value `FALSE`
    #24 11.04 -- Intel MKL-DNN compat: set DNNL_BUILD_TESTS to MKLDNN_BUILD_TESTS with value `FALSE`
    #24 11.04 -- Intel MKL-DNN compat: set DNNL_LIBRARY_TYPE to MKLDNN_LIBRARY_TYPE with value `STATIC`
    #24 11.04 -- Intel MKL-DNN compat: set DNNL_ARCH_OPT_FLAGS to MKLDNN_ARCH_OPT_FLAGS with value `-msse4`
    #24 11.04 -- Intel MKL-DNN compat: set DNNL_CPU_RUNTIME to MKLDNN_CPU_RUNTIME with value `OMP`
    #24 11.05 CMake Warning (dev) at /opt/conda/share/cmake-3.19/Modules/FindPackageHandleStandardArgs.cmake:426 (message):
    #24 11.05   The package name passed to `find_package_handle_standard_args` (OpenMP_C)
    #24 11.05   does not match the name of the calling package (OpenMP).  This can lead to
    #24 11.05   problems in calling code that expects `find_package` result variables
    #24 11.05   (e.g., `_FOUND`) to follow a certain pattern.
    #24 11.05 Call Stack (most recent call first):
    #24 11.05   cmake/Modules/FindOpenMP.cmake:576 (find_package_handle_standard_args)
    #24 11.05   third_party/ideep/mkl-dnn/cmake/OpenMP.cmake:61 (find_package)
    #24 11.05   third_party/ideep/mkl-dnn/CMakeLists.txt:119 (include)
    #24 11.05 This warning is for project developers.  Use -Wno-dev to suppress it.
    #24 11.05 
    #24 11.05 -- Found OpenMP_C: -fopenmp (found version "4.5") 
    #24 11.05 CMake Warning (dev) at /opt/conda/share/cmake-3.19/Modules/FindPackageHandleStandardArgs.cmake:426 (message):
    #24 11.05   The package name passed to `find_package_handle_standard_args` (OpenMP_CXX)
    #24 11.05   does not match the name of the calling package (OpenMP).  This can lead to
    #24 11.05   problems in calling code that expects `find_package` result variables
    #24 11.05   (e.g., `_FOUND`) to follow a certain pattern.
    #24 11.05 Call Stack (most recent call first):
    #24 11.05   cmake/Modules/FindOpenMP.cmake:576 (find_package_handle_standard_args)
    #24 11.05   third_party/ideep/mkl-dnn/cmake/OpenMP.cmake:61 (find_package)
    #24 11.05   third_party/ideep/mkl-dnn/CMakeLists.txt:119 (include)
    #24 11.05 This warning is for project developers.  Use -Wno-dev to suppress it.
    #24 11.05 
    #24 11.05 -- Found OpenMP_CXX: -fopenmp (found version "4.5") 
    #24 11.07 -- Primitive cache is enabled
    #24 11.09 -- Found MKL-DNN: TRUE
    #24 11.09 -- Looking for clock_gettime in rt
    #24 11.17 -- Looking for clock_gettime in rt - found
    #24 11.17 -- Looking for mmap
    #24 11.25 -- Looking for mmap - found
    #24 11.25 -- Looking for shm_open
    #24 11.33 -- Looking for shm_open - found
    #24 11.33 -- Looking for shm_unlink
    #24 11.41 -- Looking for shm_unlink - found
    #24 11.41 -- Looking for malloc_usable_size
    #24 11.48 -- Looking for malloc_usable_size - found
    #24 11.48 -- Performing Test C_HAS_THREAD
    #24 11.56 -- Performing Test C_HAS_THREAD - Success
    #24 11.57 -- Version: 7.0.3
    #24 11.57 -- Build type: Release
    #24 11.57 -- CXX_STANDARD: 14
    #24 11.57 -- Performing Test has_std_14_flag
    #24 11.66 -- Performing Test has_std_14_flag - Success
    #24 11.66 -- Performing Test has_std_1y_flag
    #24 11.74 -- Performing Test has_std_1y_flag - Success
    #24 11.74 -- Performing Test SUPPORTS_USER_DEFINED_LITERALS
    #24 11.83 -- Performing Test SUPPORTS_USER_DEFINED_LITERALS - Success
    #24 11.83 -- Performing Test FMT_HAS_VARIANT
    #24 11.96 -- Performing Test FMT_HAS_VARIANT - Success
    #24 11.96 -- Required features: cxx_variadic_templates
    #24 11.96 -- Looking for strtod_l
    #24 12.00 -- Looking for strtod_l - not found
    #24 12.01 -- Using Kineto with CUPTI support
    #24 12.01 -- Configuring Kineto dependency:
    #24 12.01 --   KINETO_SOURCE_DIR = /opt/pytorch/third_party/kineto/libkineto
    #24 12.01 --   KINETO_BUILD_TESTS = OFF
    #24 12.01 --   KINETO_LIBRARY_TYPE = static
    #24 12.01 --   CUDA_SOURCE_DIR = /usr/local/cuda
    #24 12.01 --   CUDA_INCLUDE_DIRS = /usr/local/cuda/include
    #24 12.01 --   CUPTI_INCLUDE_DIR = /usr/local/cuda/include
    #24 12.01 --   CUDA_cupti_LIBRARY = /usr/local/cuda/lib64/libcupti_static.a
    #24 12.01 -- Found CUPTI
    #24 12.03 -- Found PythonInterp: /opt/conda/bin/python (found version "3.9.7") 
    #24 12.09 -- Kineto: FMT_SOURCE_DIR = /opt/pytorch/third_party/fmt
    #24 12.09 -- Kineto: FMT_INCLUDE_DIR = /opt/pytorch/third_party/fmt/include
    #24 12.09 INFO CUPTI_INCLUDE_DIR = /usr/local/cuda/include
    #24 12.09 -- Configured Kineto
    #24 12.09 -- GCC 7.5.0: Adding gcc and gcc_s libs to link line
    #24 12.09 -- Performing Test HAS_WERROR_FORMAT
    #24 12.18 -- Performing Test HAS_WERROR_FORMAT - Success
    #24 12.18 -- Performing Test HAS_WERROR_CAST_FUNCTION_TYPE
    #24 12.22 -- Performing Test HAS_WERROR_CAST_FUNCTION_TYPE - Failed
    #24 12.23 -- Looking for backtrace
    #24 12.30 -- Looking for backtrace - found
    #24 12.30 -- backtrace facility detected in default set of libraries
    #24 12.30 -- Found Backtrace: /usr/include  
    #24 12.30 -- don't use NUMA
    #24 12.32 -- Performing Test COMPILER_SUPPORTS_NO_AVX256_SPLIT
    #24 12.40 -- Performing Test COMPILER_SUPPORTS_NO_AVX256_SPLIT - Success
    #24 13.54 -- Using ATen parallel backend: OMP
    #24 13.61 CMake Deprecation Warning at third_party/sleef/CMakeLists.txt:91 (cmake_policy):
    #24 13.61   The OLD behavior for policy CMP0066 will be removed from a future version
    #24 13.61   of CMake.
    #24 13.61 
    #24 13.61   The cmake-policies(7) manual explains that the OLD behaviors of all
    #24 13.61   policies are deprecated and that a policy should be set to OLD only under
    #24 13.61   specific short-term circumstances.  Projects should be ported to the NEW
    #24 13.61   behavior and not rely on setting a policy to OLD.
    #24 13.61 
    #24 13.61 
    #24 14.09 -- Found OpenSSL: /opt/conda/lib/libcrypto.so (found version "1.1.1l")  
    #24 14.10 -- Check size of long double
    #24 14.18 -- Check size of long double - done
    #24 14.19 -- Performing Test COMPILER_SUPPORTS_LONG_DOUBLE
    #24 14.27 -- Performing Test COMPILER_SUPPORTS_LONG_DOUBLE - Success
    #24 14.27 -- Performing Test COMPILER_SUPPORTS_FLOAT128
    #24 14.34 -- Performing Test COMPILER_SUPPORTS_FLOAT128 - Success
    #24 14.34 -- Performing Test COMPILER_SUPPORTS_SSE2
    #24 14.53 -- Performing Test COMPILER_SUPPORTS_SSE2 - Success
    #24 14.53 -- Performing Test COMPILER_SUPPORTS_SSE4
    #24 14.71 -- Performing Test COMPILER_SUPPORTS_SSE4 - Success
    #24 14.71 -- Performing Test COMPILER_SUPPORTS_AVX
    #24 14.89 -- Performing Test COMPILER_SUPPORTS_AVX - Success
    #24 14.89 -- Performing Test COMPILER_SUPPORTS_FMA4
    #24 15.07 -- Performing Test COMPILER_SUPPORTS_FMA4 - Success
    #24 15.07 -- Performing Test COMPILER_SUPPORTS_AVX2
    #24 15.26 -- Performing Test COMPILER_SUPPORTS_AVX2 - Success
    #24 15.26 -- Performing Test COMPILER_SUPPORTS_AVX512F
    #24 15.44 -- Performing Test COMPILER_SUPPORTS_AVX512F - Success
    #24 15.44 -- Performing Test COMPILER_SUPPORTS_OPENMP
    #24 15.53 -- Performing Test COMPILER_SUPPORTS_OPENMP - Success
    #24 15.53 -- Performing Test COMPILER_SUPPORTS_WEAK_ALIASES
    #24 15.61 -- Performing Test COMPILER_SUPPORTS_WEAK_ALIASES - Success
    #24 15.61 -- Performing Test COMPILER_SUPPORTS_BUILTIN_MATH
    #24 15.69 -- Performing Test COMPILER_SUPPORTS_BUILTIN_MATH - Success
    #24 15.69 -- Performing Test COMPILER_SUPPORTS_SYS_GETRANDOM
    #24 15.77 -- Performing Test COMPILER_SUPPORTS_SYS_GETRANDOM - Success
    #24 15.79 -- Configuring build for SLEEF-v3.6.0
    #24 15.79 -- Using option `-Wall -Wno-unused -Wno-attributes -Wno-unused-result -Wno-psabi -ffp-contract=off -fno-math-errno -fno-trapping-math` to compile libsleef
    #24 15.79 -- Building shared libs : OFF
    #24 15.79    Target system: Linux-5.4.0-88-generic
    #24 15.79    Target processor: x86_64
    #24 15.79    Host system: Linux-5.4.0-88-generic
    #24 15.79    Host processor: x86_64
    #24 15.79    Detected C compiler: GNU @ /usr/bin/cc
    #24 15.79    CMake: 3.19.6
    #24 15.79    Make program: /opt/conda/bin/ninja
    #24 15.79 -- Building static test bins: OFF
    #24 15.79 -- MPFR : LIB_MPFR-NOTFOUND
    #24 15.79 -- GMP : LIBGMP-NOTFOUND
    #24 15.79 -- RT : /usr/lib/x86_64-linux-gnu/librt.so
    #24 15.79 -- FFTW3 : LIBFFTW3-NOTFOUND
    #24 15.79 -- OPENSSL : 1.1.1l
    #24 15.79 -- SDE : SDE_COMMAND-NOTFOUND
    #24 15.79 -- RUNNING_ON_TRAVIS : 
    #24 15.79 -- COMPILER_SUPPORTS_OPENMP : 1
    #24 15.80 AT_INSTALL_INCLUDE_DIR include/ATen/core
    #24 15.80 core header install: /opt/pytorch/build/aten/src/ATen/core/TensorBody.h
    #24 15.84 -- Include NCCL operators
    #24 15.84 -- Excluding FakeLowP operators
    #24 15.85 -- Including IDEEP operators
    #24 15.85 -- Excluding image processing operators due to no opencv
    #24 15.85 -- Excluding video processing operators due to no opencv
    #24 15.85 -- MPI operators skipped due to no MPI support
    #24 15.85 -- Include Observer library
    #24 17.11 -- breakpad library not found
    #24 17.11 -- /usr/bin/c++ /opt/pytorch/torch/abi-check.cpp -o /opt/pytorch/build/abi-check
    #24 17.33 -- Determined _GLIBCXX_USE_CXX11_ABI=1
    #24 17.34 CMake Warning (dev) at torch/CMakeLists.txt:348:
    #24 17.34   Syntax Warning in cmake code at column 107
    #24 17.34 
    #24 17.34   Argument not separated from preceding token by whitespace.
    #24 17.34 This warning is for project developers.  Use -Wno-dev to suppress it.
    #24 17.34 
    #24 17.34 CMake Warning (dev) at torch/CMakeLists.txt:348:
    #24 17.34   Syntax Warning in cmake code at column 115
    #24 17.34 
    #24 17.34   Argument not separated from preceding token by whitespace.
    #24 17.34 This warning is for project developers.  Use -Wno-dev to suppress it.
    #24 17.34 
    #24 17.42 CMake Warning at cmake/public/utils.cmake:365 (message):
    #24 17.42   In the future we will require one to explicitly pass TORCH_CUDA_ARCH_LIST
    #24 17.42   to cmake instead of implicitly setting it as an env variable.  This will
    #24 17.42   become a FATAL_ERROR in future version of pytorch.
    #24 17.42 Call Stack (most recent call first):
    #24 17.42   torch/CMakeLists.txt:315 (torch_cuda_get_nvcc_gencode_flag)
    #24 17.42 
    #24 17.42 
    #24 17.42 CMake Warning (dev) at /opt/conda/share/cmake-3.19/Modules/FindPackageHandleStandardArgs.cmake:426 (message):
    #24 17.42   The package name passed to `find_package_handle_standard_args` (OpenMP_C)
    #24 17.42   does not match the name of the calling package (OpenMP).  This can lead to
    #24 17.42   problems in calling code that expects `find_package` result variables
    #24 17.42   (e.g., `_FOUND`) to follow a certain pattern.
    #24 17.42 Call Stack (most recent call first):
    #24 17.42   cmake/Modules/FindOpenMP.cmake:576 (find_package_handle_standard_args)
    #24 17.42   caffe2/CMakeLists.txt:1155 (find_package)
    #24 17.42 This warning is for project developers.  Use -Wno-dev to suppress it.
    #24 17.42 
    #24 17.42 CMake Warning (dev) at /opt/conda/share/cmake-3.19/Modules/FindPackageHandleStandardArgs.cmake:426 (message):
    #24 17.42   The package name passed to `find_package_handle_standard_args` (OpenMP_CXX)
    #24 17.42   does not match the name of the calling package (OpenMP).  This can lead to
    #24 17.42   problems in calling code that expects `find_package` result variables
    #24 17.42   (e.g., `_FOUND`) to follow a certain pattern.
    #24 17.42 Call Stack (most recent call first):
    #24 17.42   cmake/Modules/FindOpenMP.cmake:576 (find_package_handle_standard_args)
    #24 17.42   caffe2/CMakeLists.txt:1155 (find_package)
    #24 17.42 This warning is for project developers.  Use -Wno-dev to suppress it.
    #24 17.42 
    #24 17.42 -- pytorch is compiling with OpenMP. 
    #24 17.42 OpenMP CXX_FLAGS: -fopenmp. 
    #24 17.42 OpenMP libraries: /usr/lib/gcc/x86_64-linux-gnu/7/libgomp.so;/usr/lib/x86_64-linux-gnu/libpthread.so.
    #24 17.42 -- Caffe2 is compiling with OpenMP. 
    #24 17.42 OpenMP CXX_FLAGS: -fopenmp. 
    #24 17.42 OpenMP libraries: /usr/lib/gcc/x86_64-linux-gnu/7/libgomp.so;/usr/lib/x86_64-linux-gnu/libpthread.so.
    #24 17.53 -- Using lib/python3.9/site-packages as python relative installation path
    #24 17.63 CMake Warning at CMakeLists.txt:941 (message):
    #24 17.63   Generated cmake files are only fully tested if one builds with system glog,
    #24 17.63   gflags, and protobuf.  Other settings may generate files that are not well
    #24 17.63   tested.
    #24 17.63 
    #24 17.63 
    #24 17.64 -- 
    #24 17.64 -- ******** Summary ********
    #24 17.64 -- General:
    #24 17.64 --   CMake version         : 3.19.6
    #24 17.64 --   CMake command         : /opt/conda/bin/cmake
    #24 17.64 --   System                : Linux
    #24 17.64 --   C++ compiler          : /usr/bin/c++
    #24 17.64 --   C++ compiler id       : GNU
    #24 17.64 --   C++ compiler version  : 7.5.0
    #24 17.64 --   Using ccache if found : ON
    #24 17.64 --   Found ccache          : /usr/bin/ccache
    #24 17.64 --   CXX flags             :  -Wno-deprecated -fvisibility-inlines-hidden -DUSE_PTHREADPOOL -fopenmp -DNDEBUG -DUSE_KINETO -DUSE_FBGEMM -DUSE_QNNPACK -DUSE_PYTORCH_QNNPACK -DUSE_XNNPACK -DSYMBOLICATE_MOBILE_DEBUG_HANDLE -O2 -fPIC -Wno-narrowing -Wall -Wextra -Werror=return-type -Wno-missing-field-initializers -Wno-type-limits -Wno-array-bounds -Wno-unknown-pragmas -Wno-sign-compare -Wno-unused-parameter -Wno-unused-variable -Wno-unused-function -Wno-unused-result -Wno-unused-local-typedefs -Wno-strict-overflow -Wno-strict-aliasing -Wno-error=deprecated-declarations -Wno-stringop-overflow -Wno-psabi -Wno-error=pedantic -Wno-error=redundant-decls -Wno-error=old-style-cast -fdiagnostics-color=always -faligned-new -Wno-unused-but-set-variable -Wno-maybe-uninitialized -fno-math-errno -fno-trapping-math -Werror=format -Wno-stringop-overflow
    #24 17.64 --   Build type            : Release
    #24 17.64 --   Compile definitions   : TH_BLAS_MKL;ONNX_ML=1;ONNXIFI_ENABLE_EXT=1;ONNX_NAMESPACE=onnx_torch;MAGMA_V2;IDEEP_USE_MKL;HAVE_MMAP=1;_FILE_OFFSET_BITS=64;HAVE_SHM_OPEN=1;HAVE_SHM_UNLINK=1;HAVE_MALLOC_USABLE_SIZE=1;USE_EXTERNAL_MZCRC;MINIZ_DISABLE_ZIP_READER_CRC32_CHECKS
    #24 17.64 --   CMAKE_PREFIX_PATH     : /opt/conda/bin/../;/usr/local/cuda
    #24 17.64 --   CMAKE_INSTALL_PREFIX  : /opt/pytorch/torch
    #24 17.64 --   USE_GOLD_LINKER       : OFF
    #24 17.64 -- 
    #24 17.64 --   TORCH_VERSION         : 1.9.0
    #24 17.64 --   CAFFE2_VERSION        : 1.9.0
    #24 17.64 --   BUILD_CAFFE2          : ON
    #24 17.64 --   BUILD_CAFFE2_OPS      : ON
    #24 17.64 --   BUILD_CAFFE2_MOBILE   : OFF
    #24 17.64 --   BUILD_STATIC_RUNTIME_BENCHMARK: OFF
    #24 17.64 --   BUILD_TENSOREXPR_BENCHMARK: OFF
    #24 17.64 --   BUILD_BINARY          : OFF
    #24 17.64 --   BUILD_CUSTOM_PROTOBUF : ON
    #24 17.64 --     Link local protobuf : ON
    #24 17.64 --   BUILD_DOCS            : OFF
    #24 17.64 --   BUILD_PYTHON          : True
    #24 17.64 --     Python version      : 3.9.7
    #24 17.64 --     Python executable   : /opt/conda/bin/python
    #24 17.64 --     Pythonlibs version  : 3.9.7
    #24 17.64 --     Python library      : /opt/conda/lib/libpython3.9.a
    #24 17.64 --     Python includes     : /opt/conda/include/python3.9
    #24 17.64 --     Python site-packages: lib/python3.9/site-packages
    #24 17.64 --   BUILD_SHARED_LIBS     : ON
    #24 17.64 --   CAFFE2_USE_MSVC_STATIC_RUNTIME     : OFF
    #24 17.64 --   BUILD_TEST            : True
    #24 17.64 --   BUILD_JNI             : OFF
    #24 17.64 --   BUILD_MOBILE_AUTOGRAD : OFF
    #24 17.64 --   BUILD_LITE_INTERPRETER: OFF
    #24 17.64 --   INTERN_BUILD_MOBILE   : 
    #24 17.64 --   USE_BLAS              : 1
    #24 17.64 --     BLAS                : mkl
    #24 17.64 --   USE_LAPACK            : 1
    #24 17.64 --     LAPACK              : mkl
    #24 17.64 --   USE_ASAN              : OFF
    #24 17.64 --   USE_CPP_CODE_COVERAGE : OFF
    #24 17.64 --   USE_CUDA              : 1
    #24 17.64 --     Split CUDA          : OFF
    #24 17.64 --     CUDA static link    : OFF
    #24 17.64 --     USE_CUDNN           : 1
    #24 17.64 --     CUDA version        : 10.2
    #24 17.64 --     cuDNN version       : 8.2.0
    #24 17.64 --     CUDA root directory : /usr/local/cuda
    #24 17.64 --     CUDA library        : /usr/local/cuda/lib64/stubs/libcuda.so
    #24 17.64 --     cudart library      : /usr/local/cuda/lib64/libcudart.so
    #24 17.64 --     cublas library      : /usr/lib/x86_64-linux-gnu/libcublas.so
    #24 17.64 --     cufft library       : /usr/local/cuda/lib64/libcufft.so
    #24 17.64 --     curand library      : /usr/local/cuda/lib64/libcurand.so
    #24 17.64 --     cuDNN library       : /usr/lib/x86_64-linux-gnu/libcudnn.so
    #24 17.64 --     nvrtc               : /usr/local/cuda/lib64/libnvrtc.so
    #24 17.64 --     CUDA include path   : /usr/local/cuda/include
    #24 17.64 --     NVCC executable     : /usr/local/cuda/bin/nvcc
    #24 17.64 --     NVCC flags          : -Xfatbin;-compress-all;-DONNX_NAMESPACE=onnx_torch;-gencode;arch=compute_80,code=sm_80;-Xcudafe;--diag_suppress=cc_clobber_ignored,--diag_suppress=integer_sign_change,--diag_suppress=useless_using_declaration,--diag_suppress=set_but_not_used,--diag_suppress=field_without_dll_interface,--diag_suppress=base_class_has_different_dll_interface,--diag_suppress=dll_interface_conflict_none_assumed,--diag_suppress=dll_interface_conflict_dllexport_assumed,--diag_suppress=implicit_return_from_non_void_function,--diag_suppress=unsigned_compare_with_zero,--diag_suppress=declared_but_not_referenced,--diag_suppress=bad_friend_decl;-std=c++14;-Xcompiler;-fPIC;--expt-relaxed-constexpr;--expt-extended-lambda;-Wno-deprecated-gpu-targets;--expt-extended-lambda;-Xfatbin;-compress-all;-Xcompiler;-fPIC;-DCUDA_HAS_FP16=1;-D__CUDA_NO_HALF_OPERATORS__;-D__CUDA_NO_HALF_CONVERSIONS__;-D__CUDA_NO_BFLOAT16_CONVERSIONS__;-D__CUDA_NO_HALF2_OPERATORS__
    #24 17.64 --     CUDA host compiler  : /usr/bin/cc
    #24 17.64 --     NVCC --device-c     : OFF
    #24 17.64 --     USE_TENSORRT        : OFF
    #24 17.64 --   USE_ROCM              : OFF
    #24 17.64 --   USE_EIGEN_FOR_BLAS    : 
    #24 17.64 --   USE_FBGEMM            : ON
    #24 17.64 --     USE_FAKELOWP          : OFF
    #24 17.64 --   USE_KINETO            : ON
    #24 17.64 --   USE_FFMPEG            : OFF
    #24 17.64 --   USE_GFLAGS            : OFF
    #24 17.64 --   USE_GLOG              : OFF
    #24 17.64 --   USE_LEVELDB           : OFF
    #24 17.64 --   USE_LITE_PROTO        : OFF
    #24 17.64 --   USE_LMDB              : OFF
    #24 17.64 --   USE_METAL             : OFF
    #24 17.64 --   USE_PYTORCH_METAL     : OFF
    #24 17.64 --   USE_FFTW              : OFF
    #24 17.64 --   USE_MKL               : ON
    #24 17.64 --   USE_MKLDNN            : ON
    #24 17.64 --   USE_MKLDNN_ACL        : OFF
    #24 17.64 --   USE_MKLDNN_CBLAS      : OFF
    #24 17.64 --   USE_NCCL              : ON
    #24 17.64 --     USE_SYSTEM_NCCL     : OFF
    #24 17.64 --   USE_NNPACK            : ON
    #24 17.64 --   USE_NUMPY             : ON
    #24 17.64 --   USE_OBSERVERS         : ON
    #24 17.64 --   USE_OPENCL            : OFF
    #24 17.64 --   USE_OPENCV            : OFF
    #24 17.64 --   USE_OPENMP            : ON
    #24 17.64 --   USE_TBB               : OFF
    #24 17.64 --   USE_VULKAN            : OFF
    #24 17.64 --   USE_PROF              : OFF
    #24 17.64 --   USE_QNNPACK           : ON
    #24 17.64 --   USE_PYTORCH_QNNPACK   : ON
    #24 17.64 --   USE_REDIS             : OFF
    #24 17.64 --   USE_ROCKSDB           : OFF
    #24 17.64 --   USE_ZMQ               : OFF
    #24 17.64 --   USE_DISTRIBUTED       : ON
    #24 17.64 --     USE_MPI             : OFF
    #24 17.64 --     USE_GLOO            : ON
    #24 17.64 --     USE_TENSORPIPE      : ON
    #24 17.64 --   USE_DEPLOY           : OFF
    #24 17.64 --   Public Dependencies  : Threads::Threads;caffe2::mkl;caffe2::mkldnn
    #24 17.64 --   Private Dependencies : pthreadpool;cpuinfo;qnnpack;pytorch_qnnpack;nnpack;XNNPACK;fbgemm;fp16;gloo;tensorpipe;aten_op_header_gen;foxi_loader;rt;fmt::fmt-header-only;kineto;gcc_s;gcc;dl
    #24 17.65 -- Configuring done
    #24 18.73 CMake Warning at caffe2/CMakeLists.txt:791 (add_library):
    #24 18.73   Cannot generate a safe runtime search path for target torch_cpu because
    #24 18.73   files in some directories may conflict with libraries in implicit
    #24 18.73   directories:
    #24 18.73 
    #24 18.73     runtime library [libgomp.so.1] in /usr/lib/gcc/x86_64-linux-gnu/7 may be hidden by files in:
    #24 18.73       /opt/conda/lib
    #24 18.73 
    #24 18.73   Some of these libraries may not be found correctly.
    #24 18.73 
    #24 18.73 
    #24 19.13 CMake Warning at cmake/Modules_CUDA_fix/upstream/FindCUDA.cmake:1865 (add_library):
    #24 19.13   Cannot generate a safe runtime search path for target
    #24 19.13   caffe2_detectron_ops_gpu because files in some directories may conflict
    #24 19.13   with libraries in implicit directories:
    #24 19.13 
    #24 19.13     runtime library [libgomp.so.1] in /usr/lib/gcc/x86_64-linux-gnu/7 may be hidden by files in:
    #24 19.13       /opt/conda/lib
    #24 19.13 
    #24 19.13   Some of these libraries may not be found correctly.
    #24 19.13 Call Stack (most recent call first):
    #24 19.13   modules/detectron/CMakeLists.txt:13 (CUDA_ADD_LIBRARY)
    #24 19.13 
    #24 19.13 
    #24 19.15 -- Generating done
    #24 19.26 -- Build files have been written to: /opt/pytorch/build
    #24 19.35 cmake --build . --target install --config Release -- -j 64
    #24 19.59 [1/6115] Building CXX object third_party/protobuf/cmake/CMakeFiles/libprotobuf-lite.dir/__/src/google/protobuf/any_lite.cc.o
    #24 19.59 [2/6115] Building CXX object third_party/protobuf/cmake/CMakeFiles/libprotobuf-lite.dir/__/src/google/protobuf/arena.cc.o
    #24 19.60 [3/6115] Building CXX object third_party/protobuf/cmake/CMakeFiles/libprotobuf-lite.dir/__/src/google/protobuf/extension_set.cc.o
    #24 19.60 [4/6115] Building CXX object third_party/protobuf/cmake/CMakeFiles/libprotobuf-lite.dir/__/src/google/protobuf/generated_enum_util.cc.o
    #24 19.60 [5/6115] Building CXX object third_party/protobuf/cmake/CMakeFiles/libprotobuf-lite.dir/__/src/google/protobuf/generated_message_table_driven_lite.cc.o
    #24 19.60 [6/6115] Building CXX object third_party/protobuf/cmake/CMakeFiles/libprotobuf-lite.dir/__/src/google/protobuf/generated_message_util.cc.o
    #24 19.60 [7/6115] Building CXX object third_party/protobuf/cmake/CMakeFiles/libprotobuf-lite.dir/__/src/google/protobuf/implicit_weak_message.cc.o
    #24 19.60 [8/6115] Building CXX object third_party/protobuf/cmake/CMakeFiles/libprotobuf-lite.dir/__/src/google/protobuf/io/coded_stream.cc.o
    #24 19.60 [9/6115] Building CXX object third_party/protobuf/cmake/CMakeFiles/libprotobuf-lite.dir/__/src/google/protobuf/io/io_win32.cc.o
    #24 19.60 [10/6115] Building CXX object third_party/protobuf/cmake/CMakeFiles/libprotobuf-lite.dir/__/src/google/protobuf/io/strtod.cc.o
    #24 19.60 [11/6115] Building CXX object third_party/protobuf/cmake/CMakeFiles/libprotobuf-lite.dir/__/src/google/protobuf/io/zero_copy_stream.cc.o
    #24 19.61 [12/6115] Building CXX object third_party/protobuf/cmake/CMakeFiles/libprotobuf-lite.dir/__/src/google/protobuf/io/zero_copy_stream_impl.cc.o
    #24 19.61 [13/6115] Building CXX object third_party/protobuf/cmake/CMakeFiles/libprotobuf-lite.dir/__/src/google/protobuf/io/zero_copy_stream_impl_lite.cc.o
    #24 19.61 [14/6115] Building CXX object third_party/protobuf/cmake/CMakeFiles/libprotobuf-lite.dir/__/src/google/protobuf/parse_context.cc.o
    #24 19.61 [15/6115] Building CXX object third_party/protobuf/cmake/CMakeFiles/libprotobuf-lite.dir/__/src/google/protobuf/stubs/bytestream.cc.o
    #24 19.61 [16/6115] Creating directories for 'nccl_external'
    #24 19.61 [17/6115] Building CXX object third_party/protobuf/cmake/CMakeFiles/libprotobuf-lite.dir/__/src/google/protobuf/repeated_field.cc.o
    #24 19.61 [18/6115] Building CXX object third_party/protobuf/cmake/CMakeFiles/libprotobuf-lite.dir/__/src/google/protobuf/stubs/common.cc.o
    #24 19.62 [19/6115] Building CXX object third_party/protobuf/cmake/CMakeFiles/libprotobuf-lite.dir/__/src/google/protobuf/stubs/int128.cc.o
    #24 19.62 [20/6115] Building CXX object third_party/protobuf/cmake/CMakeFiles/libprotobuf-lite.dir/__/src/google/protobuf/stubs/status.cc.o
    #24 19.62 [21/6115] Building CXX object third_party/protobuf/cmake/CMakeFiles/libprotobuf-lite.dir/__/src/google/protobuf/stubs/statusor.cc.o
    #24 19.62 [22/6115] Building CXX object third_party/protobuf/cmake/CMakeFiles/libprotobuf-lite.dir/__/src/google/protobuf/stubs/stringpiece.cc.o
    #24 19.62 [23/6115] Building CXX object third_party/protobuf/cmake/CMakeFiles/libprotobuf-lite.dir/__/src/google/protobuf/stubs/stringprintf.cc.o
    #24 19.62 [24/6115] Building CXX object third_party/protobuf/cmake/CMakeFiles/libprotobuf-lite.dir/__/src/google/protobuf/stubs/structurally_valid.cc.o
    #24 19.62 [25/6115] Building CXX object third_party/protobuf/cmake/CMakeFiles/libprotobuf-lite.dir/__/src/google/protobuf/stubs/strutil.cc.o
    #24 19.63 [26/6115] Building CXX object third_party/protobuf/cmake/CMakeFiles/libprotobuf-lite.dir/__/src/google/protobuf/stubs/time.cc.o
    #24 19.63 [27/6115] Building CXX object third_party/protobuf/cmake/CMakeFiles/libprotobuf-lite.dir/__/src/google/protobuf/wire_format_lite.cc.o
    #24 19.63 [28/6115] Building CXX object third_party/protobuf/cmake/CMakeFiles/libprotobuf-lite.dir/__/src/google/protobuf/message_lite.cc.o
    #24 19.63 In file included from /usr/include/string.h:494:0,
    #24 19.63                  from ../third_party/protobuf/src/google/protobuf/stubs/port.h:39,
    #24 19.63                  from ../third_party/protobuf/src/google/protobuf/stubs/common.h:46,
    #24 19.63                  from ../third_party/protobuf/src/google/protobuf/message_lite.h:45,
    #24 19.63                  from ../third_party/protobuf/src/google/protobuf/message_lite.cc:36:
    #24 19.63 In function ‘void* memcpy(void*, const void*, size_t)’,
    #24 19.63     inlined from ‘google::protobuf::uint8* google::protobuf::io::EpsCopyOutputStream::WriteRaw(const void*, int, google::protobuf::uint8*)’ at ../third_party/protobuf/src/google/protobuf/io/coded_stream.h:699:16,
    #24 19.63     inlined from ‘bool google::protobuf::MessageLite::SerializePartialToZeroCopyStream(google::protobuf::io::ZeroCopyOutputStream*) const’ at ../third_party/protobuf/src/google/protobuf/implicit_weak_message.h:88:35,
    #24 19.63     inlined from ‘bool google::protobuf::MessageLite::SerializeToZeroCopyStream(google::protobuf::io::ZeroCopyOutputStream*) const’ at ../third_party/protobuf/src/google/protobuf/message_lite.cc:372:49:
    #24 19.63 /usr/include/x86_64-linux-gnu/bits/string_fortified.h:34:71: warning: ‘void* __builtin___memcpy_chk(void*, const void*, long unsigned int, long unsigned int)’: specified size between 18446744071562067968 and 18446744073709551615 exceeds maximum object size 9223372036854775807 [-Wstringop-overflow=]
    #24 19.63    return __builtin___memcpy_chk (__dest, __src, __len, __bos0 (__dest));
    #24 19.63                                                                        ^
    #24 19.63 In function ‘void* memcpy(void*, const void*, size_t)’,
    #24 19.63     inlined from ‘google::protobuf::uint8* google::protobuf::io::EpsCopyOutputStream::WriteRaw(const void*, int, google::protobuf::uint8*)’ at ../third_party/protobuf/src/google/protobuf/io/coded_stream.h:699:16,
    #24 19.63     inlined from ‘bool google::protobuf::MessageLite::SerializePartialToZeroCopyStream(google::protobuf::io::ZeroCopyOutputStream*) const’ at ../third_party/protobuf/src/google/protobuf/implicit_weak_message.h:88:35:
    #24 19.63 /usr/include/x86_64-linux-gnu/bits/string_fortified.h:34:71: warning: ‘void* __builtin___memcpy_chk(void*, const void*, long unsigned int, long unsigned int)’: specified size between 18446744071562067968 and 18446744073709551615 exceeds maximum object size 9223372036854775807 [-Wstringop-overflow=]
    #24 19.63    return __builtin___memcpy_chk (__dest, __src, __len, __bos0 (__dest));
    #24 19.63                                                                        ^
    #24 19.63 In function ‘void* memcpy(void*, const void*, size_t)’,
    #24 19.63     inlined from ‘google::protobuf::uint8* google::protobuf::io::EpsCopyOutputStream::WriteRaw(const void*, int, google::protobuf::uint8*)’ at ../third_party/protobuf/src/google/protobuf/io/coded_stream.h:699:16,
    #24 19.63     inlined from ‘bool google::protobuf::MessageLite::SerializePartialToZeroCopyStream(google::protobuf::io::ZeroCopyOutputStream*) const’ at ../third_party/protobuf/src/google/protobuf/implicit_weak_message.h:88:35,
    #24 19.63     inlined from ‘bool google::protobuf::MessageLite::SerializeToFileDescriptor(int) const’ at ../third_party/protobuf/src/google/protobuf/message_lite.cc:372:49:
    #24 19.63 /usr/include/x86_64-linux-gnu/bits/string_fortified.h:34:71: warning: ‘void* __builtin___memcpy_chk(void*, const void*, long unsigned int, long unsigned int)’: specified size between 18446744071562067968 and 18446744073709551615 exceeds maximum object size 9223372036854775807 [-Wstringop-overflow=]
    #24 19.63    return __builtin___memcpy_chk (__dest, __src, __len, __bos0 (__dest));
    #24 19.63                                                                        ^
    #24 19.63 In function ‘void* memcpy(void*, const void*, size_t)’,
    #24 19.63     inlined from ‘google::protobuf::uint8* google::protobuf::io::EpsCopyOutputStream::WriteRaw(const void*, int, google::protobuf::uint8*)’ at ../third_party/protobuf/src/google/protobuf/io/coded_stream.h:699:16,
    #24 19.63     inlined from ‘bool google::protobuf::MessageLite::SerializePartialToZeroCopyStream(google::protobuf::io::ZeroCopyOutputStream*) const’ at ../third_party/protobuf/src/google/protobuf/implicit_weak_message.h:88:35,
    #24 19.63     inlined from ‘bool google::protobuf::MessageLite::SerializePartialToFileDescriptor(int) const’ at ../third_party/protobuf/src/google/protobuf/message_lite.cc:401:42:
    #24 19.63 /usr/include/x86_64-linux-gnu/bits/string_fortified.h:34:71: warning: ‘void* __builtin___memcpy_chk(void*, const void*, long unsigned int, long unsigned int)’: specified size between 18446744071562067968 and 18446744073709551615 exceeds maximum object size 9223372036854775807 [-Wstringop-overflow=]
    #24 19.63    return __builtin___memcpy_chk (__dest, __src, __len, __bos0 (__dest));
    #24 19.63                                                                        ^
    #24 19.63 In function ‘void* memcpy(void*, const void*, size_t)’,
    #24 19.63     inlined from ‘google::protobuf::uint8* google::protobuf::io::EpsCopyOutputStream::WriteRaw(const void*, int, google::protobuf::uint8*)’ at ../third_party/protobuf/src/google/protobuf/io/coded_stream.h:699:16,
    #24 19.63     inlined from ‘bool google::protobuf::MessageLite::SerializePartialToZeroCopyStream(google::protobuf::io::ZeroCopyOutputStream*) const’ at ../third_party/protobuf/src/google/protobuf/implicit_weak_message.h:88:35,
    #24 19.63     inlined from ‘bool google::protobuf::MessageLite::SerializeToOstream(std::ostream*) const’ at ../third_party/protobuf/src/google/protobuf/message_lite.cc:372:49:
    #24 19.63 /usr/include/x86_64-linux-gnu/bits/string_fortified.h:34:71: warning: ‘void* __builtin___memcpy_chk(void*, const void*, long unsigned int, long unsigned int)’: specified size between 18446744071562067968 and 18446744073709551615 exceeds maximum object size 9223372036854775807 [-Wstringop-overflow=]
    #24 19.63    return __builtin___memcpy_chk (__dest, __src, __len, __bos0 (__dest));
    #24 19.63                                                                        ^
    #24 19.63 In function ‘void* memcpy(void*, const void*, size_t)’,
    #24 19.63     inlined from ‘google::protobuf::uint8* google::protobuf::io::EpsCopyOutputStream::WriteRaw(const void*, int, google::protobuf::uint8*)’ at ../third_party/protobuf/src/google/protobuf/io/coded_stream.h:699:16,
    #24 19.63     inlined from ‘bool google::protobuf::MessageLite::SerializePartialToZeroCopyStream(google::protobuf::io::ZeroCopyOutputStream*) const’ at ../third_party/protobuf/src/google/protobuf/implicit_weak_message.h:88:35,
    #24 19.63     inlined from ‘bool google::protobuf::MessageLite::SerializePartialToOstream(std::ostream*) const’ at ../third_party/protobuf/src/google/protobuf/message_lite.cc:414:60:
    #24 19.63 /usr/include/x86_64-linux-gnu/bits/string_fortified.h:34:71: warning: ‘void* __builtin___memcpy_chk(void*, const void*, long unsigned int, long unsigned int)’: specified size between 18446744071562067968 and 18446744073709551615 exceeds maximum object size 9223372036854775807 [-Wstringop-overflow=]
    #24 19.63    return __builtin___memcpy_chk (__dest, __src, __len, __bos0 (__dest));
    #24 19.63                                                                        ^
    #24 19.63 [29/6115] Building CXX object third_party/protobuf/cmake/CMakeFiles/libprotoc.dir/__/src/google/protobuf/compiler/code_generator.cc.o
    #24 19.63 [30/6115] Building CXX object third_party/protobuf/cmake/CMakeFiles/libprotoc.dir/__/src/google/protobuf/compiler/cpp/cpp_enum.cc.o
    #24 19.63 [31/6115] Building CXX object third_party/protobuf/cmake/CMakeFiles/libprotoc.dir/__/src/google/protobuf/compiler/cpp/cpp_enum_field.cc.o
    #24 19.64 [32/6115] Building CXX object third_party/protobuf/cmake/CMakeFiles/libprotoc.dir/__/src/google/protobuf/compiler/cpp/cpp_extension.cc.o
    #24 19.64 [33/6115] Building CXX object third_party/protobuf/cmake/CMakeFiles/libprotoc.dir/__/src/google/protobuf/compiler/cpp/cpp_field.cc.o
    #24 19.64 [34/6115] Building CXX object third_party/protobuf/cmake/CMakeFiles/libprotoc.dir/__/src/google/protobuf/compiler/cpp/cpp_generator.cc.o
    #24 19.64 [35/6115] Building CXX object third_party/protobuf/cmake/CMakeFiles/libprotoc.dir/__/src/google/protobuf/compiler/cpp/cpp_map_field.cc.o
    #24 19.64 [36/6115] Building CXX object third_party/protobuf/cmake/CMakeFiles/libprotoc.dir/__/src/google/protobuf/compiler/cpp/cpp_message.cc.o
    #24 19.64 [37/6115] Building CXX object third_party/protobuf/cmake/CMakeFiles/libprotoc.dir/__/src/google/protobuf/compiler/cpp/cpp_message_field.cc.o
    #24 19.64 [38/6115] Building CXX object third_party/protobuf/cmake/CMakeFiles/libprotoc.dir/__/src/google/protobuf/compiler/cpp/cpp_padding_optimizer.cc.o
    #24 19.64 [39/6115] Building CXX object third_party/protobuf/cmake/CMakeFiles/libprotoc.dir/__/src/google/protobuf/compiler/cpp/cpp_primitive_field.cc.o
    #24 19.64 [40/6115] Building CXX object third_party/protobuf/cmake/CMakeFiles/libprotoc.dir/__/src/google/protobuf/compiler/cpp/cpp_string_field.cc.o
    #24 19.64 [41/6115] Building CXX object third_party/protobuf/cmake/CMakeFiles/libprotoc.dir/__/src/google/protobuf/compiler/csharp/csharp_doc_comment.cc.o
    #24 19.65 [42/6115] Building CXX object third_party/protobuf/cmake/CMakeFiles/libprotoc.dir/__/src/google/protobuf/compiler/csharp/csharp_reflection_class.cc.o
    #24 19.65 [43/6115] Building CXX object third_party/protobuf/cmake/CMakeFiles/libprotoc.dir/__/src/google/protobuf/compiler/command_line_interface.cc.o
    #24 19.65 [44/6115] Building CXX object third_party/protobuf/cmake/CMakeFiles/libprotoc.dir/__/src/google/protobuf/compiler/cpp/cpp_file.cc.o
    #24 19.65 [45/6115] Building CXX object third_party/protobuf/cmake/CMakeFiles/libprotoc.dir/__/src/google/protobuf/compiler/cpp/cpp_helpers.cc.o
    #24 19.65 [46/6115] Building CXX object third_party/protobuf/cmake/CMakeFiles/libprotoc.dir/__/src/google/protobuf/compiler/cpp/cpp_service.cc.o
    #24 19.65 [47/6115] Building CXX object third_party/protobuf/cmake/CMakeFiles/libprotoc.dir/__/src/google/protobuf/compiler/csharp/csharp_enum.cc.o
    #24 19.65 [48/6115] Building CXX object third_party/protobuf/cmake/CMakeFiles/libprotoc.dir/__/src/google/protobuf/compiler/csharp/csharp_enum_field.cc.o
    #24 19.65 [49/6115] Building CXX object third_party/protobuf/cmake/CMakeFiles/libprotoc.dir/__/src/google/protobuf/compiler/csharp/csharp_field_base.cc.o
    #24 19.65 [50/6115] Building CXX object third_party/protobuf/cmake/CMakeFiles/libprotoc.dir/__/src/google/protobuf/compiler/csharp/csharp_generator.cc.o
    #24 19.66 [51/6115] Building CXX object third_party/protobuf/cmake/CMakeFiles/libprotoc.dir/__/src/google/protobuf/compiler/csharp/csharp_helpers.cc.o
    #24 19.66 [52/6115] Building CXX object third_party/protobuf/cmake/CMakeFiles/libprotoc.dir/__/src/google/protobuf/compiler/csharp/csharp_map_field.cc.o
    #24 19.66 [53/6115] Building CXX object third_party/protobuf/cmake/CMakeFiles/libprotoc.dir/__/src/google/protobuf/compiler/csharp/csharp_message.cc.o
    #24 19.66 [54/6115] Building CXX object third_party/protobuf/cmake/CMakeFiles/libprotoc.dir/__/src/google/protobuf/compiler/csharp/csharp_message_field.cc.o
    #24 19.66 [55/6115] Building CXX object third_party/protobuf/cmake/CMakeFiles/libprotoc.dir/__/src/google/protobuf/compiler/csharp/csharp_primitive_field.cc.o
    #24 19.66 [56/6115] Building CXX object third_party/protobuf/cmake/CMakeFiles/libprotoc.dir/__/src/google/protobuf/compiler/csharp/csharp_repeated_enum_field.cc.o
    #24 19.66 [57/6115] Building CXX object third_party/protobuf/cmake/CMakeFiles/libprotoc.dir/__/src/google/protobuf/compiler/csharp/csharp_repeated_message_field.cc.o
    #24 19.66 [58/6115] Building CXX object third_party/protobuf/cmake/CMakeFiles/libprotoc.dir/__/src/google/protobuf/compiler/csharp/csharp_repeated_primitive_field.cc.o
    #24 19.67 [59/6115] Building CXX object third_party/protobuf/cmake/CMakeFiles/libprotoc.dir/__/src/google/protobuf/compiler/csharp/csharp_source_generator_base.cc.o
    #24 19.67 [60/6115] Building CXX object third_party/protobuf/cmake/CMakeFiles/libprotoc.dir/__/src/google/protobuf/compiler/csharp/csharp_wrapper_field.cc.o
    #24 19.67 [61/6115] Building CXX object third_party/protobuf/cmake/CMakeFiles/libprotoc.dir/__/src/google/protobuf/compiler/java/java_doc_comment.cc.o
    #24 19.67 [62/6115] Building C object confu-deps/pthreadpool/CMakeFiles/pthreadpool.dir/src/pthreads.c.o
    #24 19.68 [63/6115] No download step for 'nccl_external'
    #24 19.68 [64/6115] Building CXX object third_party/protobuf/cmake/CMakeFiles/libprotoc.dir/__/src/google/protobuf/compiler/js/well_known_types_embed.cc.o
    #24 19.68 [65/6115] Building CXX object third_party/protobuf/cmake/CMakeFiles/libprotoc.dir/__/src/google/protobuf/compiler/java/java_context.cc.o
    #24 19.68 [66/6115] Building CXX object third_party/protobuf/cmake/CMakeFiles/libprotoc.dir/__/src/google/protobuf/compiler/java/java_enum.cc.o
    #24 19.68 [67/6115] Building CXX object third_party/protobuf/cmake/CMakeFiles/libprotoc.dir/__/src/google/protobuf/compiler/java/java_enum_field.cc.o
    #24 19.68 [68/6115] Building CXX object third_party/protobuf/cmake/CMakeFiles/libprotoc.dir/__/src/google/protobuf/compiler/java/java_extension_lite.cc.o
    #24 19.69 [69/6115] Building CXX object third_party/protobuf/cmake/CMakeFiles/libprotoc.dir/__/src/google/protobuf/compiler/java/java_field.cc.o
    #24 19.69 [70/6115] Building CXX object third_party/protobuf/cmake/CMakeFiles/libprotoc.dir/__/src/google/protobuf/compiler/java/java_generator.cc.o
    #24 19.69 [71/6115] Building CXX object third_party/protobuf/cmake/CMakeFiles/libprotoc.dir/__/src/google/protobuf/compiler/java/java_map_field_lite.cc.o
    #24 19.69 [72/6115] Building CXX object third_party/protobuf/cmake/CMakeFiles/libprotobuf.dir/__/src/google/protobuf/io/io_win32.cc.o
    #24 19.69 [73/6115] Building CXX object third_party/protobuf/cmake/CMakeFiles/libprotoc.dir/__/src/google/protobuf/compiler/java/java_enum_field_lite.cc.o
    #24 19.69 [74/6115] Building CXX object third_party/protobuf/cmake/CMakeFiles/libprotoc.dir/__/src/google/protobuf/compiler/java/java_enum_lite.cc.o
    #24 19.69 [75/6115] Building CXX object third_party/protobuf/cmake/CMakeFiles/libprotoc.dir/__/src/google/protobuf/compiler/java/java_extension.cc.o
    #24 19.70 [76/6115] Building CXX object third_party/protobuf/cmake/CMakeFiles/libprotoc.dir/__/src/google/protobuf/compiler/java/java_file.cc.o
    #24 19.70 [77/6115] Building CXX object third_party/protobuf/cmake/CMakeFiles/libprotoc.dir/__/src/google/protobuf/compiler/java/java_generator_factory.cc.o
    #24 19.70 [78/6115] Building CXX object third_party/protobuf/cmake/CMakeFiles/libprotoc.dir/__/src/google/protobuf/compiler/java/java_helpers.cc.o
    #24 19.70 [79/6115] Building CXX object third_party/protobuf/cmake/CMakeFiles/libprotoc.dir/__/src/google/protobuf/compiler/java/java_map_field.cc.o
    #24 19.71 [80/6115] Building CXX object third_party/protobuf/cmake/CMakeFiles/libprotoc.dir/__/src/google/protobuf/compiler/java/java_message.cc.o
    #24 19.71 [81/6115] Building CXX object third_party/protobuf/cmake/CMakeFiles/libprotoc.dir/__/src/google/protobuf/compiler/java/java_message_builder_lite.cc.o
    #24 19.71 [82/6115] Building CXX object third_party/protobuf/cmake/CMakeFiles/libprotoc.dir/__/src/google/protobuf/compiler/java/java_message_field.cc.o
    #24 19.71 [83/6115] Building CXX object third_party/protobuf/cmake/CMakeFiles/libprotoc.dir/__/src/google/protobuf/compiler/java/java_message_field_lite.cc.o
    #24 19.71 [84/6115] Building CXX object third_party/protobuf/cmake/CMakeFiles/libprotoc.dir/__/src/google/protobuf/compiler/java/java_message_lite.cc.o
    #24 19.71 [85/6115] Building CXX object third_party/protobuf/cmake/CMakeFiles/libprotoc.dir/__/src/google/protobuf/compiler/java/java_name_resolver.cc.o
    #24 19.71 [86/6115] Building CXX object third_party/protobuf/cmake/CMakeFiles/libprotoc.dir/__/src/google/protobuf/compiler/java/java_primitive_field.cc.o
    #24 19.72 [87/6115] Building CXX object third_party/protobuf/cmake/CMakeFiles/libprotoc.dir/__/src/google/protobuf/compiler/java/java_primitive_field_lite.cc.o
    #24 19.72 [88/6115] Building CXX object third_party/protobuf/cmake/CMakeFiles/libprotoc.dir/__/src/google/protobuf/compiler/java/java_shared_code_generator.cc.o
    #24 19.72 [89/6115] Building CXX object third_party/protobuf/cmake/CMakeFiles/libprotoc.dir/__/src/google/protobuf/compiler/java/java_string_field_lite.cc.o
    #24 19.72 [90/6115] No update step for 'nccl_external'
    #24 19.72 [91/6115] Building CXX object third_party/protobuf/cmake/CMakeFiles/libprotoc.dir/__/src/google/protobuf/compiler/java/java_message_builder.cc.o
    #24 19.73 [92/6115] Building CXX object third_party/protobuf/cmake/CMakeFiles/libprotoc.dir/__/src/google/protobuf/compiler/java/java_service.cc.o
    #24 19.73 [93/6115] Building CXX object third_party/protobuf/cmake/CMakeFiles/libprotoc.dir/__/src/google/protobuf/compiler/js/js_generator.cc.o
    #24 19.73 [94/6115] Linking CXX static library lib/libprotobuf-lite.a
    #24 19.73 [95/6115] Building CXX object third_party/protobuf/cmake/CMakeFiles/libprotoc.dir/__/src/google/protobuf/compiler/objectivec/objectivec_enum.cc.o
    #24 19.74 [96/6115] Building CXX object third_party/protobuf/cmake/CMakeFiles/libprotoc.dir/__/src/google/protobuf/compiler/java/java_string_field.cc.o
    #24 19.74 [97/6115] Building CXX object third_party/protobuf/cmake/CMakeFiles/libprotoc.dir/__/src/google/protobuf/compiler/objectivec/objectivec_enum_field.cc.o
    #24 19.74 [98/6115] Building CXX object third_party/protobuf/cmake/CMakeFiles/libprotoc.dir/__/src/google/protobuf/compiler/objectivec/objectivec_extension.cc.o
    #24 19.74 [99/6115] Building CXX object third_party/protobuf/cmake/CMakeFiles/libprotoc.dir/__/src/google/protobuf/compiler/objectivec/objectivec_generator.cc.o
    #24 19.74 [100/6115] Building CXX object third_party/protobuf/cmake/CMakeFiles/libprotoc.dir/__/src/google/protobuf/compiler/objectivec/objectivec_helpers.cc.o
    #24 19.74 [101/6115] Building CXX object third_party/protobuf/cmake/CMakeFiles/libprotoc.dir/__/src/google/protobuf/compiler/objectivec/objectivec_map_field.cc.o
    #24 19.74 [102/6115] Building CXX object third_party/protobuf/cmake/CMakeFiles/libprotoc.dir/__/src/google/protobuf/compiler/objectivec/objectivec_oneof.cc.o
    #24 19.75 [103/6115] Building CXX object third_party/protobuf/cmake/CMakeFiles/libprotoc.dir/__/src/google/protobuf/compiler/python/python_generator.cc.o
    #24 19.75 [104/6115] Building CXX object third_party/protobuf/cmake/CMakeFiles/libprotobuf.dir/__/src/google/protobuf/arena.cc.o
    #24 19.75 [105/6115] Building CXX object third_party/protobuf/cmake/CMakeFiles/libprotoc.dir/__/src/google/protobuf/compiler/objectivec/objectivec_field.cc.o
    #24 19.75 [106/6115] Building CXX object third_party/protobuf/cmake/CMakeFiles/libprotoc.dir/__/src/google/protobuf/compiler/objectivec/objectivec_file.cc.o
    #24 19.75 [107/6115] Building CXX object third_party/protobuf/cmake/CMakeFiles/libprotoc.dir/__/src/google/protobuf/compiler/objectivec/objectivec_message.cc.o
    #24 19.75 [108/6115] Building CXX object third_party/protobuf/cmake/CMakeFiles/libprotoc.dir/__/src/google/protobuf/compiler/objectivec/objectivec_message_field.cc.o
    #24 19.75 [109/6115] Building CXX object third_party/protobuf/cmake/CMakeFiles/libprotoc.dir/__/src/google/protobuf/compiler/objectivec/objectivec_primitive_field.cc.o
    #24 19.75 [110/6115] Building CXX object third_party/protobuf/cmake/CMakeFiles/libprotoc.dir/__/src/google/protobuf/compiler/php/php_generator.cc.o
    #24 19.75 [111/6115] Building CXX object third_party/protobuf/cmake/CMakeFiles/libprotoc.dir/__/src/google/protobuf/compiler/plugin.cc.o
    #24 19.75 [112/6115] Building CXX object third_party/protobuf/cmake/CMakeFiles/libprotoc.dir/__/src/google/protobuf/compiler/plugin.pb.cc.o
    #24 19.76 [113/6115] Building CXX object third_party/protobuf/cmake/CMakeFiles/libprotoc.dir/__/src/google/protobuf/compiler/zip_writer.cc.o
    #24 19.76 [114/6115] Building CXX object third_party/protobuf/cmake/CMakeFiles/protoc.dir/__/src/google/protobuf/compiler/main.cc.o
    #24 19.76 [115/6115] Building CXX object third_party/protobuf/cmake/CMakeFiles/libprotobuf.dir/__/src/google/protobuf/any_lite.cc.o
    #24 19.76 [116/6115] Building CXX object third_party/protobuf/cmake/CMakeFiles/libprotobuf.dir/__/src/google/protobuf/extension_set.cc.o
    #24 19.76 [117/6115] Building CXX object third_party/protobuf/cmake/CMakeFiles/libprotobuf.dir/__/src/google/protobuf/io/coded_stream.cc.o
    #24 19.76 [118/6115] Building CXX object third_party/protobuf/cmake/CMakeFiles/libprotobuf.dir/__/src/google/protobuf/io/strtod.cc.o
    #24 19.76 [119/6115] Building CXX object third_party/protobuf/cmake/CMakeFiles/libprotobuf.dir/__/src/google/protobuf/io/zero_copy_stream.cc.o
    #24 19.76 [120/6115] Building CXX object third_party/protobuf/cmake/CMakeFiles/libprotoc.dir/__/src/google/protobuf/compiler/ruby/ruby_generator.cc.o
    #24 19.77 [121/6115] Building CXX object third_party/protobuf/cmake/CMakeFiles/libprotoc.dir/__/src/google/protobuf/compiler/subprocess.cc.o
    #24 19.77 [122/6115] Building CXX object third_party/protobuf/cmake/CMakeFiles/libprotobuf.dir/__/src/google/protobuf/generated_enum_util.cc.o
    #24 19.77 [123/6115] Building CXX object third_party/protobuf/cmake/CMakeFiles/libprotobuf.dir/__/src/google/protobuf/generated_message_table_driven_lite.cc.o
    #24 19.77 [124/6115] Building CXX object third_party/protobuf/cmake/CMakeFiles/libprotobuf.dir/__/src/google/protobuf/generated_message_util.cc.o
    #24 19.77 [125/6115] Building CXX object third_party/protobuf/cmake/CMakeFiles/libprotobuf.dir/__/src/google/protobuf/implicit_weak_message.cc.o
    #24 19.78 [126/6115] Building CXX object third_party/protobuf/cmake/CMakeFiles/libprotobuf.dir/__/src/google/protobuf/io/zero_copy_stream_impl.cc.o
    #24 19.78 [127/6115] Building CXX object third_party/protobuf/cmake/CMakeFiles/libprotobuf.dir/__/src/google/protobuf/io/zero_copy_stream_impl_lite.cc.o
    #24 19.79 [128/6115] Building CXX object third_party/protobuf/cmake/CMakeFiles/libprotobuf.dir/__/src/google/protobuf/parse_context.cc.o
    #24 19.79 [129/6115] Building CXX object third_party/protobuf/cmake/CMakeFiles/libprotobuf.dir/__/src/google/protobuf/repeated_field.cc.o
    #24 19.79 [130/6115] Building CXX object third_party/protobuf/cmake/CMakeFiles/libprotobuf.dir/__/src/google/protobuf/stubs/bytestream.cc.o
    #24 19.80 [131/6115] Building CXX object third_party/protobuf/cmake/CMakeFiles/libprotobuf.dir/__/src/google/protobuf/stubs/common.cc.o
    #24 19.80 [132/6115] Building CXX object third_party/protobuf/cmake/CMakeFiles/libprotobuf.dir/__/src/google/protobuf/stubs/int128.cc.o
    #24 19.80 [133/6115] Building CXX object third_party/protobuf/cmake/CMakeFiles/libprotobuf.dir/__/src/google/protobuf/stubs/status.cc.o
    #24 19.80 [134/6115] Building CXX object third_party/protobuf/cmake/CMakeFiles/libprotobuf.dir/__/src/google/protobuf/stubs/statusor.cc.o
    #24 19.80 [135/6115] Building CXX object third_party/protobuf/cmake/CMakeFiles/libprotobuf.dir/__/src/google/protobuf/stubs/stringpiece.cc.o
    #24 19.80 [136/6115] Building CXX object third_party/protobuf/cmake/CMakeFiles/libprotobuf.dir/__/src/google/protobuf/stubs/stringprintf.cc.o
    #24 19.80 [137/6115] Building CXX object third_party/protobuf/cmake/CMakeFiles/libprotobuf.dir/__/src/google/protobuf/stubs/structurally_valid.cc.o
    #24 19.80 [138/6115] Building CXX object third_party/protobuf/cmake/CMakeFiles/libprotobuf.dir/__/src/google/protobuf/stubs/strutil.cc.o
    #24 19.81 [139/6115] Building CXX object third_party/protobuf/cmake/CMakeFiles/libprotobuf.dir/__/src/google/protobuf/stubs/time.cc.o
    #24 19.82 [140/6115] Building CXX object third_party/protobuf/cmake/CMakeFiles/libprotobuf.dir/__/src/google/protobuf/wire_format_lite.cc.o
    #24 19.84 [141/6115] No patch step for 'nccl_external'
    #24 19.85 [142/6115] Building CXX object third_party/protobuf/cmake/CMakeFiles/libprotobuf.dir/__/src/google/protobuf/io/gzip_stream.cc.o
    #24 19.85 [143/6115] Building CXX object third_party/protobuf/cmake/CMakeFiles/libprotobuf.dir/__/src/google/protobuf/message_lite.cc.o
    #24 19.85 In file included from /usr/include/string.h:494:0,
    #24 19.85                  from ../third_party/protobuf/src/google/protobuf/stubs/port.h:39,
    #24 19.85                  from ../third_party/protobuf/src/google/protobuf/stubs/common.h:46,
    #24 19.85                  from ../third_party/protobuf/src/google/protobuf/message_lite.h:45,
    #24 19.85                  from ../third_party/protobuf/src/google/protobuf/message_lite.cc:36:
    #24 19.85 In function ‘void* memcpy(void*, const void*, size_t)’,
    #24 19.85     inlined from ‘google::protobuf::uint8* google::protobuf::io::EpsCopyOutputStream::WriteRaw(const void*, int, google::protobuf::uint8*)’ at ../third_party/protobuf/src/google/protobuf/io/coded_stream.h:699:16,
    #24 19.85     inlined from ‘bool google::protobuf::MessageLite::SerializePartialToZeroCopyStream(google::protobuf::io::ZeroCopyOutputStream*) const’ at ../third_party/protobuf/src/google/protobuf/implicit_weak_message.h:88:35,
    #24 19.85     inlined from ‘bool google::protobuf::MessageLite::SerializeToZeroCopyStream(google::protobuf::io::ZeroCopyOutputStream*) const’ at ../third_party/protobuf/src/google/protobuf/message_lite.cc:372:49:
    #24 19.85 /usr/include/x86_64-linux-gnu/bits/string_fortified.h:34:71: warning: ‘void* __builtin___memcpy_chk(void*, const void*, long unsigned int, long unsigned int)’: specified size between 18446744071562067968 and 18446744073709551615 exceeds maximum object size 9223372036854775807 [-Wstringop-overflow=]
    #24 19.85    return __builtin___memcpy_chk (__dest, __src, __len, __bos0 (__dest));
    #24 19.85                                                                        ^
    #24 19.85 In function ‘void* memcpy(void*, const void*, size_t)’,
    #24 19.85     inlined from ‘google::protobuf::uint8* google::protobuf::io::EpsCopyOutputStream::WriteRaw(const void*, int, google::protobuf::uint8*)’ at ../third_party/protobuf/src/google/protobuf/io/coded_stream.h:699:16,
    #24 19.85     inlined from ‘bool google::protobuf::MessageLite::SerializePartialToZeroCopyStream(google::protobuf::io::ZeroCopyOutputStream*) const’ at ../third_party/protobuf/src/google/protobuf/implicit_weak_message.h:88:35:
    #24 19.85 /usr/include/x86_64-linux-gnu/bits/string_fortified.h:34:71: warning: ‘void* __builtin___memcpy_chk(void*, const void*, long unsigned int, long unsigned int)’: specified size between 18446744071562067968 and 18446744073709551615 exceeds maximum object size 9223372036854775807 [-Wstringop-overflow=]
    #24 19.85    return __builtin___memcpy_chk (__dest, __src, __len, __bos0 (__dest));
    #24 19.85                                                                        ^
    #24 19.85 In function ‘void* memcpy(void*, const void*, size_t)’,
    #24 19.85     inlined from ‘google::protobuf::uint8* google::protobuf::io::EpsCopyOutputStream::WriteRaw(const void*, int, google::protobuf::uint8*)’ at ../third_party/protobuf/src/google/protobuf/io/coded_stream.h:699:16,
    #24 19.85     inlined from ‘bool google::protobuf::MessageLite::SerializePartialToZeroCopyStream(google::protobuf::io::ZeroCopyOutputStream*) const’ at ../third_party/protobuf/src/google/protobuf/implicit_weak_message.h:88:35,
    #24 19.85     inlined from ‘bool google::protobuf::MessageLite::SerializeToFileDescriptor(int) const’ at ../third_party/protobuf/src/google/protobuf/message_lite.cc:372:49:
    #24 19.85 /usr/include/x86_64-linux-gnu/bits/string_fortified.h:34:71: warning: ‘void* __builtin___memcpy_chk(void*, const void*, long unsigned int, long unsigned int)’: specified size between 18446744071562067968 and 18446744073709551615 exceeds maximum object size 9223372036854775807 [-Wstringop-overflow=]
    #24 19.85    return __builtin___memcpy_chk (__dest, __src, __len, __bos0 (__dest));
    #24 19.85                                                                        ^
    #24 19.85 In function ‘void* memcpy(void*, const void*, size_t)’,
    #24 19.85     inlined from ‘google::protobuf::uint8* google::protobuf::io::EpsCopyOutputStream::WriteRaw(const void*, int, google::protobuf::uint8*)’ at ../third_party/protobuf/src/google/protobuf/io/coded_stream.h:699:16,
    #24 19.85     inlined from ‘bool google::protobuf::MessageLite::SerializePartialToZeroCopyStream(google::protobuf::io::ZeroCopyOutputStream*) const’ at ../third_party/protobuf/src/google/protobuf/implicit_weak_message.h:88:35,
    #24 19.85     inlined from ‘bool google::protobuf::MessageLite::SerializePartialToFileDescriptor(int) const’ at ../third_party/protobuf/src/google/protobuf/message_lite.cc:401:42:
    #24 19.85 /usr/include/x86_64-linux-gnu/bits/string_fortified.h:34:71: warning: ‘void* __builtin___memcpy_chk(void*, const void*, long unsigned int, long unsigned int)’: specified size between 18446744071562067968 and 18446744073709551615 exceeds maximum object size 9223372036854775807 [-Wstringop-overflow=]
    #24 19.85    return __builtin___memcpy_chk (__dest, __src, __len, __bos0 (__dest));
    #24 19.85                                                                        ^
    #24 19.85 In function ‘void* memcpy(void*, const void*, size_t)’,
    #24 19.85     inlined from ‘google::protobuf::uint8* google::protobuf::io::EpsCopyOutputStream::WriteRaw(const void*, int, google::protobuf::uint8*)’ at ../third_party/protobuf/src/google/protobuf/io/coded_stream.h:699:16,
    #24 19.85     inlined from ‘bool google::protobuf::MessageLite::SerializePartialToZeroCopyStream(google::protobuf::io::ZeroCopyOutputStream*) const’ at ../third_party/protobuf/src/google/protobuf/implicit_weak_message.h:88:35,
    #24 19.85     inlined from ‘bool google::protobuf::MessageLite::SerializeToOstream(std::ostream*) const’ at ../third_party/protobuf/src/google/protobuf/message_lite.cc:372:49:
    #24 19.85 /usr/include/x86_64-linux-gnu/bits/string_fortified.h:34:71: warning: ‘void* __builtin___memcpy_chk(void*, const void*, long unsigned int, long unsigned int)’: specified size between 18446744071562067968 and 18446744073709551615 exceeds maximum object size 9223372036854775807 [-Wstringop-overflow=]
    #24 19.85    return __builtin___memcpy_chk (__dest, __src, __len, __bos0 (__dest));
    #24 19.85                                                                        ^
    #24 19.85 In function ‘void* memcpy(void*, const void*, size_t)’,
    #24 19.85     inlined from ‘google::protobuf::uint8* google::protobuf::io::EpsCopyOutputStream::WriteRaw(const void*, int, google::protobuf::uint8*)’ at ../third_party/protobuf/src/google/protobuf/io/coded_stream.h:699:16,
    #24 19.85     inlined from ‘bool google::protobuf::MessageLite::SerializePartialToZeroCopyStream(google::protobuf::io::ZeroCopyOutputStream*) const’ at ../third_party/protobuf/src/google/protobuf/implicit_weak_message.h:88:35,
    #24 19.85     inlined from ‘bool google::protobuf::MessageLite::SerializePartialToOstream(std::ostream*) const’ at ../third_party/protobuf/src/google/protobuf/message_lite.cc:414:60:
    #24 19.85 /usr/include/x86_64-linux-gnu/bits/string_fortified.h:34:71: warning: ‘void* __builtin___memcpy_chk(void*, const void*, long unsigned int, long unsigned int)’: specified size between 18446744071562067968 and 18446744073709551615 exceeds maximum object size 9223372036854775807 [-Wstringop-overflow=]
    #24 19.85    return __builtin___memcpy_chk (__dest, __src, __len, __bos0 (__dest));
    #24 19.85                                                                        ^
    #24 19.87 [144/6115] No configure step for 'nccl_external'
    #24 19.93 [145/6115] Building C object confu-deps/NNPACK/CMakeFiles/nnpack_reference_layers.dir/src/ref/convolution-input-gradient.c.o
    #24 19.94 [146/6115] Building C object confu-deps/NNPACK/CMakeFiles/nnpack_reference_layers.dir/src/ref/convolution-output.c.o
    #24 19.94 [147/6115] Building C object confu-deps/cpuinfo/CMakeFiles/cpuinfo_internals.dir/src/cache.c.o
    #24 19.96 [148/6115] Building C object confu-deps/cpuinfo/CMakeFiles/cpuinfo_internals.dir/src/x86/vendor.c.o
    #24 19.98 [149/6115] Building C object confu-deps/cpuinfo/CMakeFiles/cpuinfo_internals.dir/src/x86/info.c.o
    #24 20.01 [150/6115] Building C object confu-deps/cpuinfo/CMakeFiles/cpuinfo_internals.dir/src/init.c.o
    #24 20.06 [151/6115] Building C object confu-deps/cpuinfo/CMakeFiles/cpuinfo_internals.dir/src/x86/uarch.c.o
    #24 20.06 [152/6115] Building C object confu-deps/pthreadpool/CMakeFiles/pthreadpool.dir/src/memory.c.o
    #24 20.08 [153/6115] Building C object confu-deps/cpuinfo/CMakeFiles/cpuinfo_internals.dir/src/x86/init.c.o
    #24 20.08 [154/6115] Building C object confu-deps/pthreadpool/CMakeFiles/pthreadpool.dir/src/legacy-api.c.o
    #24 20.11 [155/6115] Building C object confu-deps/cpuinfo/CMakeFiles/cpuinfo_internals.dir/src/api.c.o
    #24 20.13 [156/6115] Building C object confu-deps/cpuinfo/CMakeFiles/cpuinfo_internals.dir/src/x86/topology.c.o
    #24 20.13 [157/6115] Building C object confu-deps/cpuinfo/CMakeFiles/cpuinfo_internals.dir/src/x86/isa.c.o
    #24 20.16 [158/6115] Building C object confu-deps/cpuinfo/CMakeFiles/cpuinfo_internals.dir/src/x86/name.c.o
    #24 20.19 [159/6115] Building C object confu-deps/cpuinfo/CMakeFiles/cpuinfo_internals.dir/src/x86/cache/descriptor.c.o
    #24 20.21 [160/6115] Building C object confu-deps/cpuinfo/CMakeFiles/cpuinfo_internals.dir/src/x86/cache/init.c.o
    #24 20.27 [161/6115] Building C object confu-deps/cpuinfo/CMakeFiles/cpuinfo_internals.dir/src/x86/cache/deterministic.c.o
    #24 20.35 [162/6115] Building C object confu-deps/cpuinfo/CMakeFiles/cpuinfo_internals.dir/src/x86/linux/cpuinfo.c.o
    #24 20.40 [163/6115] Building C object confu-deps/cpuinfo/CMakeFiles/cpuinfo_internals.dir/src/x86/linux/init.c.o
    #24 20.42 [164/6115] Building C object confu-deps/cpuinfo/CMakeFiles/cpuinfo_internals.dir/src/linux/smallfile.c.o
    #24 20.44 [165/6115] Building C object confu-deps/cpuinfo/CMakeFiles/cpuinfo_internals.dir/src/linux/multiline.c.o
    #24 20.44 [166/6115] Building C object confu-deps/cpuinfo/CMakeFiles/cpuinfo.dir/src/init.c.o
    #24 20.44 [167/6115] Building C object confu-deps/cpuinfo/CMakeFiles/cpuinfo.dir/src/cache.c.o
    #24 20.45 [168/6115] Building C object confu-deps/cpuinfo/CMakeFiles/cpuinfo_internals.dir/src/linux/cpulist.c.o
    #24 20.46 [169/6115] Building C object confu-deps/cpuinfo/CMakeFiles/cpuinfo_internals.dir/src/linux/processors.c.o
    #24 20.48 [170/6115] Building C object confu-deps/cpuinfo/CMakeFiles/cpuinfo.dir/src/x86/info.c.o
    #24 20.49 [171/6115] Building C object confu-deps/pthreadpool/CMakeFiles/pthreadpool.dir/src/fastpath.c.o
    #24 20.50 [172/6115] Building C object confu-deps/cpuinfo/CMakeFiles/cpuinfo.dir/src/x86/init.c.o
    #24 20.51 [173/6115] Building C object confu-deps/cpuinfo/CMakeFiles/cpuinfo.dir/src/api.c.o
    #24 20.52 [174/6115] Building CXX object third_party/protobuf/cmake/CMakeFiles/libprotobuf.dir/__/src/google/protobuf/service.cc.o
    #24 20.57 [175/6115] Building C object confu-deps/cpuinfo/CMakeFiles/cpuinfo.dir/src/x86/vendor.c.o
    #24 20.58 [176/6115] Building C object confu-deps/cpuinfo/CMakeFiles/cpuinfo.dir/src/x86/topology.c.o
    #24 20.60 [177/6115] Building C object confu-deps/cpuinfo/CMakeFiles/cpuinfo.dir/src/x86/uarch.c.o
    #24 20.71 [178/6115] Building C object confu-deps/cpuinfo/CMakeFiles/cpuinfo.dir/src/x86/cache/descriptor.c.o
    #24 20.73 [179/6115] Building CXX object third_party/protobuf/cmake/CMakeFiles/libprotobuf.dir/__/src/google/protobuf/stubs/substitute.cc.o
    #24 20.74 [180/6115] Building CXX object third_party/protobuf/cmake/CMakeFiles/libprotobuf.dir/__/src/google/protobuf/util/internal/json_escaping.cc.o
    #24 20.78 [181/6115] Building CXX object third_party/protobuf/cmake/CMakeFiles/libprotobuf.dir/__/src/google/protobuf/util/internal/error_listener.cc.o
    #24 20.80 [182/6115] Building C object confu-deps/cpuinfo/CMakeFiles/cpuinfo.dir/src/x86/cache/deterministic.c.o
    #24 20.82 [183/6115] Building C object confu-deps/cpuinfo/CMakeFiles/cpuinfo.dir/src/x86/name.c.o
    #24 20.84 [184/6115] Building C object confu-deps/cpuinfo/CMakeFiles/cpuinfo.dir/src/x86/isa.c.o
    #24 20.84 [185/6115] Building C object confu-deps/cpuinfo/CMakeFiles/cpuinfo.dir/src/x86/cache/init.c.o
    #24 20.85 [186/6115] Building C object confu-deps/cpuinfo/CMakeFiles/cpuinfo.dir/src/x86/linux/cpuinfo.c.o
    #24 20.85 [187/6115] Building C object confu-deps/cpuinfo/CMakeFiles/cpuinfo.dir/src/linux/smallfile.c.o
    #24 20.85 [188/6115] Building C object confu-deps/cpuinfo/CMakeFiles/cpuinfo.dir/src/linux/multiline.c.o
    #24 20.86 [189/6115] Building C object confu-deps/cpuinfo/CMakeFiles/cpuinfo.dir/src/linux/cpulist.c.o
    #24 20.89 [190/6115] Building C object confu-deps/cpuinfo/deps/clog/CMakeFiles/clog.dir/src/clog.c.o
    #24 20.89 ../third_party/cpuinfo/deps/clog/src/clog.c: In function ‘clog_vlog_fatal’:
    #24 20.89 ../third_party/cpuinfo/deps/clog/src/clog.c:112:4: warning: ignoring return value of ‘write’, declared with attribute warn_unused_result [-Wunused-result]
    #24 20.89     write(STDERR_FILENO, out_buffer, prefix_chars + format_chars + CLOG_SUFFIX_LENGTH);
    #24 20.89     ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
    #24 20.89 ../third_party/cpuinfo/deps/clog/src/clog.c: In function ‘clog_vlog_error’:
    #24 20.89 ../third_party/cpuinfo/deps/clog/src/clog.c:188:4: warning: ignoring return value of ‘write’, declared with attribute warn_unused_result [-Wunused-result]
    #24 20.89     write(STDERR_FILENO, out_buffer, prefix_chars + format_chars + CLOG_SUFFIX_LENGTH);
    #24 20.89     ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
    #24 20.89 ../third_party/cpuinfo/deps/clog/src/clog.c: In function ‘clog_vlog_warning’:
    #24 20.89 ../third_party/cpuinfo/deps/clog/src/clog.c:264:4: warning: ignoring return value of ‘write’, declared with attribute warn_unused_result [-Wunused-result]
    #24 20.89     write(STDERR_FILENO, out_buffer, prefix_chars + format_chars + CLOG_SUFFIX_LENGTH);
    #24 20.89     ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
    #24 20.89 ../third_party/cpuinfo/deps/clog/src/clog.c: In function ‘clog_vlog_info’:
    #24 20.89 ../third_party/cpuinfo/deps/clog/src/clog.c:340:4: warning: ignoring return value of ‘write’, declared with attribute warn_unused_result [-Wunused-result]
    #24 20.89     write(STDOUT_FILENO, out_buffer, prefix_chars + format_chars + CLOG_SUFFIX_LENGTH);
    #24 20.89     ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
    #24 20.89 ../third_party/cpuinfo/deps/clog/src/clog.c: In function ‘clog_vlog_debug’:
    #24 20.89 ../third_party/cpuinfo/deps/clog/src/clog.c:416:4: warning: ignoring return value of ‘write’, declared with attribute warn_unused_result [-Wunused-result]
    #24 20.89     write(STDOUT_FILENO, out_buffer, prefix_chars + format_chars + CLOG_SUFFIX_LENGTH);
    #24 20.89     ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
    #24 20.92 [191/6115] Building C object confu-deps/cpuinfo/CMakeFiles/cpuinfo.dir/src/linux/processors.c.o
    #24 20.95 [192/6115] Building CXX object third_party/protobuf/cmake/CMakeFiles/libprotobuf.dir/__/src/google/protobuf/util/delimited_message_util.cc.o
    #24 20.97 [193/6115] Building C object confu-deps/QNNPACK/CMakeFiles/qnnpack.dir/src/init.c.o
    #24 20.98 [194/6115] Building CXX object third_party/protobuf/cmake/CMakeFiles/libprotobuf.dir/__/src/google/protobuf/any.cc.o
    #24 20.99 [195/6115] Building C object confu-deps/cpuinfo/CMakeFiles/cpuinfo.dir/src/x86/linux/init.c.o
    #24 21.00 [196/6115] Building C object confu-deps/QNNPACK/CMakeFiles/qnnpack.dir/src/add.c.o
    #24 21.02 [197/6115] Building C object confu-deps/QNNPACK/CMakeFiles/qnnpack.dir/src/global-average-pooling.c.o
    #24 21.04 [198/6115] Linking C static library lib/libclog.a
    #24 21.05 [199/6115] Building C object confu-deps/QNNPACK/CMakeFiles/qnnpack.dir/src/average-pooling.c.o
    #24 21.05 [200/6115] Building C object confu-deps/QNNPACK/CMakeFiles/qnnpack.dir/src/channel-shuffle.c.o
    #24 21.06 [201/6115] Building C object confu-deps/QNNPACK/CMakeFiles/qnnpack.dir/src/clamp.c.o
    #24 21.07 [202/6115] Building C object confu-deps/QNNPACK/CMakeFiles/qnnpack.dir/src/leaky-relu.c.o
    #24 21.09 [203/6115] Building C object confu-deps/QNNPACK/CMakeFiles/qnnpack.dir/src/max-pooling.c.o
    #24 21.11 [204/6115] Linking C static library lib/libcpuinfo_internals.a
    #24 21.13 [205/6115] Building C object confu-deps/QNNPACK/CMakeFiles/qnnpack.dir/src/fully-connected.c.o
    #24 21.14 [206/6115] Building C object confu-deps/QNNPACK/CMakeFiles/qnnpack.dir/src/sigmoid.c.o
    #24 21.15 [207/6115] Linking C static library lib/libcpuinfo.a
    #24 21.15 [208/6115] Building C object confu-deps/QNNPACK/CMakeFiles/qnnpack.dir/src/operator-delete.c.o
    #24 21.16 [209/6115] Building C object confu-deps/QNNPACK/CMakeFiles/qnnpack.dir/src/u8lut32norm/scalar.c.o
    #24 21.16 [210/6115] Building C object confu-deps/QNNPACK/CMakeFiles/qnnpack.dir/src/deconvolution.c.o
    #24 21.17 [211/6115] Building CXX object third_party/protobuf/cmake/CMakeFiles/libprotobuf.dir/__/src/google/protobuf/empty.pb.cc.o
    #24 21.19 [212/6115] Building C object confu-deps/QNNPACK/CMakeFiles/qnnpack.dir/src/convolution.c.o
    #24 21.19 [213/6115] Building C object confu-deps/QNNPACK/CMakeFiles/qnnpack.dir/src/sgemm/6x8-psimd.c.o
    #24 21.21 [214/6115] Building C object confu-deps/QNNPACK/CMakeFiles/qnnpack.dir/src/softargmax.c.o
    #24 21.21 [215/6115] Building CXX object third_party/protobuf/cmake/CMakeFiles/libprotobuf.dir/__/src/google/protobuf/duration.pb.cc.o
    #24 21.21 [216/6115] Building CXX object third_party/protobuf/cmake/CMakeFiles/libprotobuf.dir/__/src/google/protobuf/io/printer.cc.o
    #24 21.22 [217/6115] Building C object confu-deps/QNNPACK/CMakeFiles/qnnpack.dir/src/x8lut/scalar.c.o
    #24 21.23 [218/6115] Building CXX object third_party/protobuf/cmake/CMakeFiles/libprotobuf.dir/__/src/google/protobuf/field_mask.pb.cc.o
    #24 21.24 [219/6115] Building CXX object third_party/protobuf/cmake/CMakeFiles/libprotobuf.dir/__/src/google/protobuf/util/internal/object_writer.cc.o
    #24 21.28 [220/6115] Building CXX object third_party/protobuf/cmake/CMakeFiles/libprotobuf.dir/__/src/google/protobuf/timestamp.pb.cc.o
    #24 21.29 [221/6115] Building C object confu-deps/pthreadpool/CMakeFiles/pthreadpool.dir/src/portable-api.c.o
    #24 21.31 [222/6115] Building C object confu-deps/QNNPACK/CMakeFiles/qnnpack.dir/src/indirection.c.o
    #24 21.31 [223/6115] Building CXX object third_party/protobuf/cmake/CMakeFiles/libprotobuf.dir/__/src/google/protobuf/source_context.pb.cc.o
    #24 21.32 [224/6115] Building C object confu-deps/QNNPACK/CMakeFiles/qnnpack.dir/src/q8avgpool/up8x9-sse2.c.o
    #24 21.34 [225/6115] Building C object confu-deps/QNNPACK/CMakeFiles/qnnpack.dir/src/q8avgpool/up8xm-sse2.c.o
    #24 21.35 [226/6115] Building CXX object third_party/protobuf/cmake/CMakeFiles/libprotobuf.dir/__/src/google/protobuf/any.pb.cc.o
    #24 21.35 [227/6115] Building C object confu-deps/QNNPACK/CMakeFiles/qnnpack.dir/src/u8clamp/sse2.c.o
    #24 21.36 [228/6115] Building C object confu-deps/QNNPACK/CMakeFiles/qnnpack.dir/src/x8zip/x2-sse2.c.o
    #24 21.37 [229/6115] Building CXX object third_party/protobuf/cmake/CMakeFiles/libprotobuf.dir/__/src/google/protobuf/util/field_comparator.cc.o
    #24 21.38 [230/6115] Building C object confu-deps/QNNPACK/CMakeFiles/qnnpack.dir/src/operator-run.c.o
    #24 21.38 [231/6115] Building C object confu-deps/QNNPACK/CMakeFiles/qnnpack.dir/src/q8gavgpool/up8x7-sse2.c.o
    #24 21.39 [232/6115] Building C object confu-deps/QNNPACK/CMakeFiles/qnnpack.dir/src/q8gavgpool/up8xm-sse2.c.o
    #24 21.40 [233/6115] Building C object confu-deps/QNNPACK/CMakeFiles/qnnpack.dir/src/u8maxpool/sub16-sse2.c.o
    #24 21.40 [234/6115] Building C object confu-deps/QNNPACK/CMakeFiles/qnnpack.dir/src/u8rmax/sse2.c.o
    #24 21.42 [235/6115] Linking C static library lib/libpthreadpool.a
    #24 21.43 [236/6115] Building CXX object third_party/protobuf/cmake/CMakeFiles/libprotobuf.dir/__/src/google/protobuf/util/internal/type_info_test_helper.cc.o
    #24 21.44 [237/6115] Building C object confu-deps/QNNPACK/CMakeFiles/qnnpack.dir/src/q8avgpool/mp8x9p8q-sse2.c.o
    #24 21.44 [238/6115] Building C object confu-deps/QNNPACK/CMakeFiles/qnnpack.dir/src/q8gavgpool/mp8x7p7q-sse2.c.o
    #24 21.45 [239/6115] Building C object confu-deps/QNNPACK/CMakeFiles/qnnpack.dir/src/u8maxpool/16x9p8q-sse2.c.o
    #24 21.46 [240/6115] Building C object confu-deps/QNNPACK/CMakeFiles/qnnpack.dir/src/x8zip/x3-sse2.c.o
    #24 21.50 [241/6115] Building C object confu-deps/QNNPACK/CMakeFiles/qnnpack.dir/src/x8zip/x4-sse2.c.o
    #24 21.50 [242/6115] Building CXX object third_party/protobuf/cmake/CMakeFiles/libprotobuf.dir/__/src/google/protobuf/io/tokenizer.cc.o
    #24 21.52 [243/6115] Building CXX object third_party/protobuf/cmake/CMakeFiles/libprotobuf.dir/__/src/google/protobuf/util/internal/field_mask_utility.cc.o
    #24 21.54 [244/6115] Building CXX object third_party/protobuf/cmake/CMakeFiles/libprotobuf.dir/__/src/google/protobuf/util/time_util.cc.o
    #24 21.56 [245/6115] Building C object confu-deps/QNNPACK/CMakeFiles/qnnpack.dir/src/x8zip/xm-sse2.c.o
    #24 21.58 [246/6115] Building C object confu-deps/pytorch_qnnpack/CMakeFiles/pytorch_qnnpack.dir/src/channel-shuffle.c.o
    #24 21.60 [247/6115] Building CXX object third_party/protobuf/cmake/CMakeFiles/libprotobuf.dir/__/src/google/protobuf/util/internal/json_objectwriter.cc.o
    #24 21.61 [248/6115] Building C object confu-deps/QNNPACK/CMakeFiles/qnnpack.dir/src/q8conv/4x4c2-sse2.c.o
    #24 21.61 [249/6115] Building C object confu-deps/pytorch_qnnpack/CMakeFiles/pytorch_qnnpack.dir/src/init.c.o
    #24 21.61 [250/6115] Building C object confu-deps/pytorch_qnnpack/CMakeFiles/pytorch_qnnpack.dir/src/add.c.o
    #24 21.61 [251/6115] Building C object confu-deps/pytorch_qnnpack/CMakeFiles/pytorch_qnnpack.dir/src/average-pooling.c.o
    #24 21.62 [252/6115] Building C object confu-deps/pytorch_qnnpack/CMakeFiles/pytorch_qnnpack.dir/src/fully-connected.c.o
    #24 21.63 [253/6115] Building C object confu-deps/pytorch_qnnpack/CMakeFiles/pytorch_qnnpack.dir/src/global-average-pooling.c.o
    #24 21.64 [254/6115] Building CXX object third_party/protobuf/cmake/CMakeFiles/libprotobuf.dir/__/src/google/protobuf/util/internal/json_stream_parser.cc.o
    #24 21.64 [255/6115] Building C object confu-deps/QNNPACK/CMakeFiles/qnnpack.dir/src/q8gemm/4x4c2-sse2.c.o
    #24 21.66 [256/6115] Building C object confu-deps/QNNPACK/CMakeFiles/qnnpack.dir/src/q8vadd/sse2.c.o
    #24 21.66 [257/6115] Building C object confu-deps/pytorch_qnnpack/CMakeFiles/pytorch_qnnpack.dir/src/clamp.c.o
    #24 21.67 [258/6115] Building C object confu-deps/pytorch_qnnpack/CMakeFiles/pytorch_qnnpack.dir/src/hardsigmoid.c.o
    #24 21.67 [259/6115] Building C object confu-deps/pytorch_qnnpack/CMakeFiles/pytorch_qnnpack.dir/src/hardswish.c.o
    #24 21.68 [260/6115] Building C object confu-deps/pytorch_qnnpack/CMakeFiles/pytorch_qnnpack.dir/src/sigmoid.c.o
    #24 21.68 [261/6115] Building C object confu-deps/QNNPACK/CMakeFiles/qnnpack.dir/src/q8dwconv/up8x9-sse2.c.o
    #24 21.70 [262/6115] Building C object confu-deps/QNNPACK/CMakeFiles/qnnpack.dir/src/q8gemm/2x4c8-sse2.c.o
    #24 21.72 [263/6115] Building C object confu-deps/pytorch_qnnpack/CMakeFiles/pytorch_qnnpack.dir/src/deconvolution.c.o
    #24 21.74 [264/6115] Building C object confu-deps/pytorch_qnnpack/CMakeFiles/pytorch_qnnpack.dir/src/fully-connected-sparse.c.o
    #24 21.75 [265/6115] Building C object confu-deps/pytorch_qnnpack/CMakeFiles/pytorch_qnnpack.dir/src/leaky-relu.c.o
    #24 21.76 [266/6115] Building C object confu-deps/pytorch_qnnpack/CMakeFiles/pytorch_qnnpack.dir/src/max-pooling.c.o
    #24 21.76 [267/6115] Building C object confu-deps/pytorch_qnnpack/CMakeFiles/pytorch_qnnpack.dir/src/softargmax.c.o
    #24 21.77 [268/6115] Building CXX object third_party/protobuf/cmake/CMakeFiles/libprotobuf.dir/__/src/google/protobuf/reflection_ops.cc.o
    #24 21.77 [269/6115] Building CXX object third_party/protobuf/cmake/CMakeFiles/libprotobuf.dir/__/src/google/protobuf/util/json_util.cc.o
    #24 21.77 [270/6115] Building C object confu-deps/QNNPACK/CMakeFiles/qnnpack.dir/src/q8dwconv/mp8x25-sse2.c.o
    #24 21.77 [271/6115] Building C object confu-deps/pytorch_qnnpack/CMakeFiles/pytorch_qnnpack.dir/src/convolution.c.o
    #24 21.78 [272/6115] Building C object confu-deps/pytorch_qnnpack/CMakeFiles/pytorch_qnnpack.dir/src/tanh.c.o
    #24 21.81 [273/6115] Building C object confu-deps/pytorch_qnnpack/CMakeFiles/pytorch_qnnpack.dir/src/operator-delete.c.o
    #24 21.83 [274/6115] Building C object confu-deps/pytorch_qnnpack/CMakeFiles/pytorch_qnnpack.dir/src/x8lut/scalar.c.o
    #24 21.84 [275/6115] Building C object confu-deps/pytorch_qnnpack/CMakeFiles/pytorch_qnnpack.dir/src/u8lut32norm/scalar.c.o
    #24 21.84 [276/6115] Building C object confu-deps/pytorch_qnnpack/CMakeFiles/pytorch_qnnpack.dir/src/sgemm/6x8-psimd.c.o
    #24 21.85 [277/6115] Building C object confu-deps/pytorch_qnnpack/CMakeFiles/pytorch_qnnpack.dir/src/q8avgpool/up8xm-sse2.c.o
    #24 21.86 [278/6115] Building C object confu-deps/pytorch_qnnpack/CMakeFiles/pytorch_qnnpack.dir/src/q8avgpool/up8x9-sse2.c.o
    #24 21.88 [279/6115] Linking C static library lib/libqnnpack.a
    #24 21.89 [280/6115] Building CXX object third_party/protobuf/cmake/CMakeFiles/libprotobuf.dir/__/src/google/protobuf/util/internal/utility.cc.o
    #24 21.90 [281/6115] Building CXX object third_party/protobuf/cmake/CMakeFiles/libprotobuf.dir/__/src/google/protobuf/util/internal/type_info.cc.o
    #24 21.91 [282/6115] Building C object confu-deps/pytorch_qnnpack/CMakeFiles/pytorch_qnnpack.dir/src/indirection.c.o
    #24 21.91 [283/6115] Building C object confu-deps/pytorch_qnnpack/CMakeFiles/pytorch_qnnpack.dir/src/q8avgpool/mp8x9p8q-sse2.c.o
    #24 21.92 [284/6115] Building C object confu-deps/pytorch_qnnpack/CMakeFiles/pytorch_qnnpack.dir/src/q8gavgpool/up8xm-sse2.c.o
    #24 21.93 [285/6115] Building CXX object third_party/protobuf/cmake/CMakeFiles/libprotobuf.dir/__/src/google/protobuf/dynamic_message.cc.o
    #24 21.94 [286/6115] Building C object confu-deps/pytorch_qnnpack/CMakeFiles/pytorch_qnnpack.dir/src/q8gavgpool/up8x7-sse2.c.o
    #24 21.94 [287/6115] Building CXX object third_party/protobuf/cmake/CMakeFiles/libprotobuf.dir/__/src/google/protobuf/unknown_field_set.cc.o
    #24 21.95 [288/6115] Building C object confu-deps/pytorch_qnnpack/CMakeFiles/pytorch_qnnpack.dir/src/q8gavgpool/mp8x7p7q-sse2.c.o
    #24 21.95 [289/6115] Building CXX object third_party/protobuf/cmake/CMakeFiles/libprotobuf.dir/__/src/google/protobuf/extension_set_heavy.cc.o
    #24 21.95 [290/6115] Building CXX object confu-deps/pytorch_qnnpack/CMakeFiles/pytorch_qnnpack.dir/src/fc-prepack.cc.o
    #24 21.97 [291/6115] Building CXX object third_party/protobuf/cmake/CMakeFiles/libprotobuf.dir/__/src/google/protobuf/compiler/importer.cc.o
    #24 21.97 [292/6115] Building CXX object third_party/protobuf/cmake/CMakeFiles/libprotobuf.dir/__/src/google/protobuf/api.pb.cc.o
    #24 21.99 [293/6115] Building CXX object confu-deps/pytorch_qnnpack/CMakeFiles/pytorch_qnnpack.dir/src/conv-prepack.cc.o
    #24 22.02 [294/6115] Building C object confu-deps/pytorch_qnnpack/CMakeFiles/pytorch_qnnpack.dir/src/u8clamp/sse2.c.o
    #24 22.02 [295/6115] Building C object confu-deps/pytorch_qnnpack/CMakeFiles/pytorch_qnnpack.dir/src/operator-run.c.o
    #24 22.03 [296/6115] Building C object confu-deps/pytorch_qnnpack/CMakeFiles/pytorch_qnnpack.dir/src/u8maxpool/sub16-sse2.c.o
    #24 22.03 [297/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/math/roundz-scalar-trunc.c.o
    #24 22.03 [298/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/math/sigmoid-scalar-rr2-lut64-p2-div.c.o
    #24 22.04 [299/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/x32-unpool/scalar.c.o
    #24 22.04 [300/6115] Building C object confu-deps/pytorch_qnnpack/CMakeFiles/pytorch_qnnpack.dir/src/u8maxpool/16x9p8q-sse2.c.o
    #24 22.04 [301/6115] Building C object confu-deps/pytorch_qnnpack/CMakeFiles/pytorch_qnnpack.dir/src/u8rmax/sse2.c.o
    #24 22.04 [302/6115] Building C object confu-deps/pytorch_qnnpack/CMakeFiles/pytorch_qnnpack.dir/src/x8zip/x2-sse2.c.o
    #24 22.04 [303/6115] Building C object confu-deps/pytorch_qnnpack/CMakeFiles/pytorch_qnnpack.dir/src/x8zip/x3-sse2.c.o
    #24 22.04 [304/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/qs8-requantization/fp32-scalar-lrintf.c.o
    #24 22.04 [305/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/x32-zip/x2-scalar.c.o
    #24 22.05 [306/6115] Building CXX object third_party/protobuf/cmake/CMakeFiles/libprotobuf.dir/__/src/google/protobuf/util/field_mask_util.cc.o
    #24 22.05 [307/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/math/sigmoid-scalar-rr2-lut2048-p1-div.c.o
    #24 22.05 [308/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/x32-pad/scalar-int.c.o
    #24 22.06 [309/6115] Building C object confu-deps/pytorch_qnnpack/CMakeFiles/pytorch_qnnpack.dir/src/x8zip/x4-sse2.c.o
    #24 22.06 [310/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/math/sigmoid-scalar-rr2-p5-div.c.o
    #24 22.06 [311/6115] Building C object confu-deps/NNPACK/CMakeFiles/nnpack_reference_layers.dir/src/ref/fully-connected-output.c.o
    #24 22.06 [312/6115] Building C object confu-deps/NNPACK/CMakeFiles/nnpack_reference_layers.dir/src/ref/convolution-kernel.c.o
    #24 22.07 [313/6115] Building C object confu-deps/NNPACK/CMakeFiles/nnpack_reference_layers.dir/src/ref/softmax-output.c.o
    #24 22.07 [314/6115] Building C object confu-deps/NNPACK/CMakeFiles/nnpack_reference_layers.dir/src/ref/max-pooling-output.c.o
    #24 22.08 [315/6115] Building C object confu-deps/pytorch_qnnpack/CMakeFiles/pytorch_qnnpack.dir/src/x8zip/xm-sse2.c.o
    #24 22.09 [316/6115] Building C object confu-deps/NNPACK/CMakeFiles/nnpack_reference_layers.dir/src/ref/relu-output.c.o
    #24 22.10 [317/6115] Building C object confu-deps/NNPACK/CMakeFiles/nnpack_reference_layers.dir/src/ref/relu-input-gradient.c.o
    #24 22.11 [318/6115] Building CXX object confu-deps/pytorch_qnnpack/CMakeFiles/pytorch_qnnpack.dir/src/fc-dynamic-run.cc.o
    #24 22.12 [319/6115] Building C object confu-deps/pytorch_qnnpack/CMakeFiles/pytorch_qnnpack.dir/src/q8gemm/2x4c8-sse2.c.o
    #24 22.14 [320/6115] Linking C static library lib/libnnpack_reference_layers.a
    #24 22.15 [321/6115] Building CXX object confu-deps/pytorch_qnnpack/CMakeFiles/pytorch_qnnpack.dir/src/fc-run.cc.o
    #24 22.17 [322/6115] Building C object confu-deps/pytorch_qnnpack/CMakeFiles/pytorch_qnnpack.dir/src/q8conv/4x4c2-sse2.c.o
    #24 22.19 [323/6115] Building CXX object third_party/protobuf/cmake/CMakeFiles/libprotobuf.dir/__/src/google/protobuf/util/type_resolver_util.cc.o
    #24 22.20 [324/6115] Building C object confu-deps/pytorch_qnnpack/CMakeFiles/pytorch_qnnpack.dir/src/q8dwconv/up8x9-sse2-per-channel.c.o
    #24 22.22 [325/6115] Building CXX object third_party/protobuf/cmake/CMakeFiles/libprotobuf.dir/__/src/google/protobuf/wrappers.pb.cc.o
    #24 22.22 [326/6115] Building C object confu-deps/pytorch_qnnpack/CMakeFiles/pytorch_qnnpack.dir/src/q8gemm/4x4c2-sse2.c.o
    #24 22.22 [327/6115] Building C object confu-deps/pytorch_qnnpack/CMakeFiles/pytorch_qnnpack.dir/src/q8vadd/sse2.c.o
    #24 22.22 [328/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/operators/channel-shuffle-nc.c.o
    #24 22.23 [329/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/operators/depth-to-space-nchw2nhwc.c.o
    #24 22.23 [330/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/operators/depth-to-space-nhwc.c.o
    #24 22.23 [331/6115] Building C object confu-deps/pytorch_qnnpack/CMakeFiles/pytorch_qnnpack.dir/src/q8dwconv/up8x9-sse2.c.o
    #24 22.23 [332/6115] Building C object confu-deps/pytorch_qnnpack/CMakeFiles/pytorch_qnnpack.dir/src/q8gemm_sparse/8x4-packA-sse2.c.o
    #24 22.23 [333/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/operators/argmax-pooling-nhwc.c.o
    #24 22.24 [334/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/subgraph/add2.c.o
    #24 22.25 [335/6115] Building CXX object confu-deps/pytorch_qnnpack/CMakeFiles/pytorch_qnnpack.dir/src/deconv-run.cc.o
    #24 22.25 [336/6115] Building C object confu-deps/pytorch_qnnpack/CMakeFiles/pytorch_qnnpack.dir/src/q8gemm/4x4c2-dq-sse2.c.o
    #24 22.26 [337/6115] Building C object confu-deps/pytorch_qnnpack/CMakeFiles/pytorch_qnnpack.dir/src/q8gemm_sparse/8x4c1x4-dq-packedA-sse2.c.o
    #24 22.26 [338/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/operators/leaky-relu-nc.c.o
    #24 22.26 [339/6115] Building CXX object third_party/protobuf/cmake/CMakeFiles/libprotobuf.dir/__/src/google/protobuf/util/internal/proto_writer.cc.o
    #24 22.28 [340/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/operators/prelu-nc.c.o
    #24 22.29 [341/6115] Building C object confu-deps/pytorch_qnnpack/CMakeFiles/pytorch_qnnpack.dir/src/q8dwconv/mp8x25-sse2.c.o
    #24 22.29 [342/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/operators/constant-pad-nd.c.o
    #24 22.29 [343/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/operators/global-average-pooling-ncw.c.o
    #24 22.29 [344/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/operators/resize-bilinear-nchw.c.o
    #24 22.29 [345/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/operators/resize-bilinear-nhwc.c.o
    #24 22.30 [346/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/operators/sigmoid-nc.c.o
    #24 22.30 [347/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/operators/unpooling-nhwc.c.o
    #24 22.30 [348/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/subgraph/argmax-pooling-2d.c.o
    #24 22.30 [349/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/subgraph/average-pooling-2d.c.o
    #24 22.30 [350/6115] Building C object confu-deps/pytorch_qnnpack/CMakeFiles/pytorch_qnnpack.dir/src/q8dwconv/mp8x25-sse2-per-channel.c.o
    #24 22.31 [351/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/operators/fully-connected-nc.c.o
    #24 22.31 [352/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/operators/max-pooling-nhwc.c.o
    #24 22.32 [353/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/operators/softmax-nc.c.o
    #24 22.32 [354/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/subgraph/abs.c.o
    #24 22.32 [355/6115] Building CXX object third_party/protobuf/cmake/CMakeFiles/libprotobuf.dir/__/src/google/protobuf/message.cc.o
    #24 22.33 [356/6115] Building CXX object confu-deps/pytorch_qnnpack/CMakeFiles/pytorch_qnnpack.dir/src/conv-run.cc.o
    #24 22.33 [357/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/operators/binary-elementwise-nd.c.o
    #24 22.35 [358/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/operators/global-average-pooling-nwc.c.o
    #24 22.35 [359/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/subgraph/bankers-rounding.c.o
    #24 22.35 [360/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/subgraph/ceiling.c.o
    #24 22.36 [361/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/operators/average-pooling-nhwc.c.o
    #24 22.36 [362/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/operators/convolution-nchw.c.o
    #24 22.37 [363/6115] Building CXX object third_party/protobuf/cmake/CMakeFiles/libprotobuf.dir/__/src/google/protobuf/util/internal/protostream_objectsource.cc.o
    #24 22.38 [364/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/operators/deconvolution-nhwc.c.o
    #24 22.39 [365/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/subgraph/divide.c.o
    #24 22.39 [366/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/subgraph/depth-to-space.c.o
    #24 22.39 [367/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/subgraph/clamp.c.o
    #24 22.40 [368/6115] Building CXX object confu-deps/pytorch_qnnpack/CMakeFiles/pytorch_qnnpack.dir/src/pack_block_sparse.cc.o
    #24 22.40 [369/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/subgraph/convolution-2d.c.o
    #24 22.40 [370/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/subgraph/deconvolution-2d.c.o
    #24 22.41 [371/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/subgraph/minimum2.c.o
    #24 22.41 [372/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/tables/exp2-k-over-64.c.o
    #24 22.41 [373/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/subgraph/elu.c.o
    #24 22.41 [374/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/subgraph/floor.c.o
    #24 22.42 [375/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/subgraph/fully-connected.c.o
    #24 22.42 [376/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/tables/exp2-k-over-2048.c.o
    #24 22.42 [377/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/operators/unary-elementwise-nc.c.o
    #24 22.43 [378/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/subgraph/depthwise-convolution-2d.c.o
    #24 22.43 [379/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/subgraph/hardswish.c.o
    #24 22.43 [380/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/tables/exp2minus-k-over-8.c.o
    #24 22.43 [381/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/subgraph/global-average-pooling-2d.c.o
    #24 22.43 [382/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/subgraph/max-pooling-2d.c.o
    #24 22.43 [383/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/subgraph/maximum2.c.o
    #24 22.44 [384/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/tables/exp2minus-k-over-64.c.o
    #24 22.44 [385/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/subgraph/leaky-relu.c.o
    #24 22.45 [386/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/tables/exp2minus-k-over-4.c.o
    #24 22.45 [387/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/tables/exp2minus-k-over-16.c.o
    #24 22.45 [388/6115] Building CXX object third_party/protobuf/cmake/CMakeFiles/libprotobuf.dir/__/src/google/protobuf/util/internal/datapiece.cc.o
    #24 22.45 [389/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/subgraph/negate.c.o
    #24 22.45 [390/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/subgraph/prelu.c.o
    #24 22.45 [391/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/memory.c.o
    #24 22.46 [392/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/tables/exp2minus-k-over-2048.c.o
    #24 22.46 [393/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/subgraph/multiply2.c.o
    #24 22.46 [394/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/subgraph/softmax.c.o
    #24 22.47 [395/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/subgraph/squared-difference.c.o
    #24 22.47 [396/6115] Linking CXX static library lib/libpytorch_qnnpack.a
    #24 22.48 [397/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/subgraph/sigmoid.c.o
    #24 22.48 [398/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/operator-delete.c.o
    #24 22.49 [399/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/subgraph/unpooling-2d.c.o
    #24 22.49 [400/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/operators/convolution-nhwc.c.o
    #24 22.49 [401/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/operator-strings.c.o
    #24 22.49 [402/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/f32-argmaxpool/4x-scalar-c1.c.o
    #24 22.50 [403/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/subgraph/static-resize-bilinear-2d.c.o
    #24 22.50 [404/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/subgraph/square-root.c.o
    #24 22.50 [405/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/subgraph/subtract.c.o
    #24 22.50 [406/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/subgraph-strings.c.o
    #24 22.51 [407/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/subgraph/static-constant-pad.c.o
    #24 22.51 [408/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/subgraph/square.c.o
    #24 22.51 [409/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/f32-clamp/gen/scalar-x4.c.o
    #24 22.52 [410/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/f32-clamp/gen/scalar-x2.c.o
    #24 22.53 [411/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/f32-clamp/gen/scalar-x1.c.o
    #24 22.54 [412/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/subgraph/static-reshape.c.o
    #24 22.54 [413/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/tensor.c.o
    #24 22.54 [414/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/f32-argmaxpool/9x-scalar-c1.c.o
    #24 22.54 [415/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/f32-avgpool/9x-minmax-scalar-c1.c.o
    #24 22.55 [416/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/f32-dwconv/gen/up1x4-minmax-scalar-acc2.c.o
    #24 22.55 [417/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/f32-dwconv/gen/up1x4-minmax-scalar.c.o
    #24 22.55 [418/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/memory-planner.c.o
    #24 22.55 [419/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/f32-argmaxpool/9p8x-scalar-c1.c.o
    #24 22.55 [420/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/f32-dwconv/gen/up1x4-scalar-acc2.c.o
    #24 22.56 [421/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/f32-dwconv/gen/up1x9-minmax-scalar-acc2.c.o
    #24 22.56 [422/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/f32-dwconv/gen/up1x4-scalar.c.o
    #24 22.56 [423/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/f32-dwconv/gen/up1x9-scalar.c.o
    #24 22.57 [424/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/f32-dwconv/gen/up1x9-scalar-acc2.c.o
    #24 22.57 [425/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/f32-dwconv/gen/up2x4-scalar.c.o
    #24 22.58 [426/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/f32-avgpool/9p8x-minmax-scalar-c1.c.o
    #24 22.58 [427/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/f32-dwconv/gen/up2x4-minmax-scalar.c.o
    #24 22.58 [428/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/f32-dwconv/gen/up2x4-scalar-acc2.c.o
    #24 22.58 [429/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/f32-dwconv/gen/up1x9-minmax-scalar.c.o
    #24 22.59 [430/6115] Building CXX object third_party/protobuf/cmake/CMakeFiles/libprotobuf.dir/__/src/google/protobuf/util/internal/default_value_objectwriter.cc.o
    #24 22.59 [431/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/f32-dwconv2d-chw/gen/3x3p1-minmax-scalar-1x1-acc4.c.o
    #24 22.59 [432/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/f32-conv-hwc/3x3s2p1c3x4-scalar-1x1.c.o
    #24 22.60 [433/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/f32-dwconv/gen/up1x25-scalar.c.o
    #24 22.60 [434/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/f32-dwconv2d-chw/gen/3x3p1-minmax-scalar-1x1-acc3.c.o
    #24 22.60 [435/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/f32-dwconv2d-chw/gen/3x3p1-minmax-scalar-1x1.c.o
    #24 22.60 [436/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/f32-dwconv/gen/up1x25-minmax-scalar.c.o
    #24 22.60 [437/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/f32-dwconv/gen/up1x25-scalar-acc2.c.o
    #24 22.60 [438/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/f32-dwconv/gen/up2x9-minmax-scalar.c.o
    #24 22.61 [439/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/f32-dwconv/gen/up2x9-minmax-scalar-acc2.c.o
    #24 22.61 [440/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/f32-dwconv2d-chw/gen/3x3p1-minmax-scalar-1x1-acc2.c.o
    #24 22.61 [441/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/f32-conv-hwc2chw/3x3s2p1c3x4-scalar-1x1.c.o
    #24 22.61 [442/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/f32-dwconv/gen/up2x4-minmax-scalar-acc2.c.o
    #24 22.61 [443/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/f32-dwconv/gen/up2x9-scalar.c.o
    #24 22.62 [444/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/f32-dwconv/gen/up2x9-scalar-acc2.c.o
    #24 22.62 [445/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/f32-conv-hwc/3x3s2p0p1c3x4-scalar-1x1.c.o
    #24 22.62 [446/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/f32-dwconv2d-chw/gen/3x3p1-minmax-scalar-2x1.c.o
    #24 22.63 [447/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/f32-dwconv/gen/up2x25-scalar-acc2.c.o
    #24 22.64 [448/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/f32-dwconv2d-chw/gen/3x3p1-minmax-scalar-2x1-acc2.c.o
    #24 22.65 [449/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/f32-dwconv/gen/up2x25-minmax-scalar.c.o
    #24 22.65 [450/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/f32-dwconv2d-chw/gen/3x3p1-minmax-scalar-3x1.c.o
    #24 22.66 [451/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/f32-dwconv2d-chw/gen/3x3s2p1-minmax-scalar-2x1-acc2.c.o
    #24 22.66 [452/6115] Building CXX object third_party/protobuf/cmake/CMakeFiles/libprotobuf.dir/__/src/google/protobuf/type.pb.cc.o
    #24 22.66 [453/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/f32-dwconv2d-chw/gen/3x3s2p1-minmax-scalar-1x1-acc3.c.o
    #24 22.66 [454/6115] Building CXX object third_party/protobuf/cmake/CMakeFiles/libprotobuf.dir/__/src/google/protobuf/map_field.cc.o
    #24 22.66 [455/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/subgraph.c.o
    #24 22.66 [456/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/f32-dwconv2d-chw/gen/3x3p1-minmax-scalar-4x1.c.o
    #24 22.66 [457/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/f32-dwconv2d-chw/gen/3x3s2p1-minmax-scalar-1x1.c.o
    #24 22.67 [458/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/f32-dwconv2d-chw/gen/3x3s2p1-minmax-scalar-1x1-acc2.c.o
    #24 22.67 [459/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/f32-dwconv/gen/up1x25-minmax-scalar-acc2.c.o
    #24 22.67 [460/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/f32-dwconv/gen/up2x25-minmax-scalar-acc2.c.o
    #24 22.68 [461/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/f32-dwconv2d-chw/gen/3x3p1-minmax-scalar-5x1.c.o
    #24 22.68 [462/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/f32-dwconv2d-chw/gen/3x3s2p1-minmax-scalar-2x1.c.o
    #24 22.69 [463/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/f32-dwconv2d-chw/gen/3x3s2p1-minmax-scalar-3x1.c.o
    #24 22.69 [464/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/indirection.c.o
    #24 22.69 [465/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/f32-dwconv/gen/up2x25-scalar.c.o
    #24 22.69 [466/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/f32-dwconv2d-chw/gen/3x3p1-minmax-scalar-6x1.c.o
    #24 22.70 [467/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/f32-dwconv2d-chw/gen/3x3s2p1-minmax-scalar-1x1-acc4.c.o
    #24 22.70 [468/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/f32-dwconv2d-chw/gen/3x3s2p1-minmax-scalar-4x1.c.o
    #24 22.70 [469/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/f32-dwconv2d-chw/gen/5x5p2-minmax-scalar-1x1-acc2.c.o
    #24 22.70 [470/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/f32-dwconv2d-chw/gen/5x5p2-minmax-scalar-1x1-acc3.c.o
    #24 22.70 [471/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/f32-dwconv2d-chw/gen/5x5p2-minmax-scalar-1x1-acc4.c.o
    #24 22.70 [472/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/init.c.o
    #24 22.70 [473/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/f32-dwconv2d-chw/gen/5x5p2-minmax-scalar-1x1-acc5.c.o
    #24 22.70 [474/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/f32-dwconv2d-chw/gen/5x5p2-minmax-scalar-1x1.c.o
    #24 22.72 [475/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/operator-run.c.o
    #24 22.72 [476/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/f32-dwconv2d-chw/gen/5x5s2p2-minmax-scalar-1x1-acc5.c.o
    #24 22.72 [477/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/f32-gavgpool-cw/scalar-x1.c.o
    #24 22.72 [478/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/f32-gavgpool/7x-minmax-scalar-c1.c.o
    #24 22.72 [479/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/f32-gavgpool/7p7x-minmax-scalar-c1.c.o
    #24 22.72 [480/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/f32-dwconv2d-chw/gen/5x5p2-minmax-scalar-2x1.c.o
    #24 22.73 [481/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/f32-gemm/gen-inc/2x4inc-minmax-scalar.c.o
    #24 22.73 [482/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/f32-dwconv2d-chw/gen/5x5s2p2-minmax-scalar-1x1-acc2.c.o
    #24 22.73 [483/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/f32-dwconv2d-chw/gen/5x5s2p2-minmax-scalar-1x1-acc3.c.o
    #24 22.74 [484/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/f32-dwconv2d-chw/gen/5x5p2-minmax-scalar-2x1-acc2.c.o
    #24 22.74 [485/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/f32-dwconv2d-chw/gen/5x5p2-minmax-scalar-2x1-acc3.c.o
    #24 22.74 [486/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/f32-dwconv2d-chw/gen/5x5s2p2-minmax-scalar-1x1-acc4.c.o
    #24 22.74 [487/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/f32-dwconv2d-chw/gen/5x5s2p2-minmax-scalar-1x1.c.o
    #24 22.74 [488/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/f32-gemm/gen-inc/1x4inc-minmax-scalar.c.o
    #24 22.75 [489/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/f32-gemm/gen/1x4-minmax-scalar.c.o
    #24 22.75 [490/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/f32-gemm/gen/1x4-relu-scalar.c.o
    #24 22.76 [491/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/f32-dwconv2d-chw/gen/5x5p2-minmax-scalar-3x1.c.o
    #24 22.76 [492/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/f32-dwconv2d-chw/gen/5x5s2p2-minmax-scalar-2x1-acc3.c.o
    #24 22.76 [493/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/f32-dwconv2d-chw/gen/5x5s2p2-minmax-scalar-2x1.c.o
    #24 22.76 [494/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/f32-gemm/gen-inc/4x4inc-minmax-scalar.c.o
    #24 22.76 [495/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/f32-gemm/gen/2x4-minmax-scalar.c.o
    #24 22.76 [496/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/f32-gemm/gen/2x4-relu-scalar.c.o
    #24 22.76 [497/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/f32-gemm/gen/2x4-scalar.c.o
    #24 22.76 [498/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/f32-hswish/gen/hswish-scalar-x1.c.o
    #24 22.76 [499/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/f32-dwconv2d-chw/gen/5x5p2-minmax-scalar-3x1-acc2.c.o
    #24 22.76 [500/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/f32-dwconv2d-chw/gen/5x5s2p2-minmax-scalar-2x1-acc2.c.o
    #24 22.77 [501/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/f32-gemm/gen/1x4-scalar.c.o
    #24 22.77 [502/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/f32-gemm/gen/4x2-minmax-scalar.c.o
    #24 22.77 [503/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/f32-ibilinear-chw/gen/scalar-p1.c.o
    #24 22.78 [504/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/f32-gemm/gen/4x2-relu-scalar.c.o
    #24 22.78 [505/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/f32-gemm/gen/4x2-scalar.c.o
    #24 22.78 [506/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/f32-hswish/gen/hswish-scalar-x2.c.o
    #24 22.78 [507/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/f32-dwconv2d-chw/gen/5x5s2p2-minmax-scalar-3x1.c.o
    #24 22.79 [508/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/f32-gemm/gen/4x4-relu-scalar.c.o
    #24 22.79 [509/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/f32-gemm/gen/4x4-scalar.c.o
    #24 22.79 [510/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/f32-hswish/gen/hswish-scalar-x4.c.o
    #24 22.79 [511/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/f32-ibilinear-chw/gen/scalar-p2.c.o
    #24 22.79 [512/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/f32-ibilinear/gen/scalar-c1.c.o
    #24 22.79 [513/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/f32-ibilinear/gen/scalar-c2.c.o
    #24 22.79 [514/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/f32-dwconv2d-chw/gen/5x5s2p2-minmax-scalar-3x1-acc2.c.o
    #24 22.79 [515/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/f32-gemm/gen/4x4-minmax-scalar.c.o
    #24 22.80 [516/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/f32-ibilinear-chw/gen/scalar-p4.c.o
    #24 22.80 [517/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/f32-ibilinear/gen/scalar-c4.c.o
    #24 22.80 [518/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/f32-igemm/gen/1x4-minmax-scalar.c.o
    #24 22.80 [519/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/f32-igemm/gen/1x4-relu-scalar.c.o
    #24 22.81 [520/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/f32-igemm/gen/1x4-scalar.c.o
    #24 22.81 [521/6115] Building CXX object third_party/protobuf/cmake/CMakeFiles/libprotobuf.dir/__/src/google/protobuf/util/internal/protostream_objectwriter.cc.o
    #24 22.82 [522/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/f32-igemm/gen/4x2-minmax-scalar.c.o
    #24 22.82 [523/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/f32-igemm/gen/2x4-minmax-scalar.c.o
    #24 22.83 [524/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/f32-igemm/gen/2x4-relu-scalar.c.o
    #24 22.83 [525/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/f32-igemm/gen/2x4-scalar.c.o
    #24 22.83 [526/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/f32-igemm/gen/4x2-relu-scalar.c.o
    #24 22.83 [527/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/f32-igemm/gen/4x2-scalar.c.o
    #24 22.83 [528/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/f32-ppmm/gen/2x4-minmax-scalar.c.o
    #24 22.83 [529/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/f32-pavgpool/9x-minmax-scalar-c1.c.o
    #24 22.84 [530/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/f32-prelu/gen/scalar-2x1.c.o
    #24 22.84 [531/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/f32-igemm/gen/4x4-minmax-scalar.c.o
    #24 22.84 [532/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/f32-igemm/gen/4x4-scalar.c.o
    #24 22.85 [533/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/f32-maxpool/9p8x-minmax-scalar-c1.c.o
    #24 22.85 [534/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/f32-ppmm/gen/4x2-minmax-scalar.c.o
    #24 22.85 [535/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/runtime.c.o
    #24 22.85 [536/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/f32-igemm/gen/4x4-relu-scalar.c.o
    #24 22.85 [537/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/f32-ppmm/gen/3x3-minmax-scalar.c.o
    #24 22.86 [538/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/f32-prelu/gen/scalar-2x4.c.o
    #24 22.86 [539/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/f32-raddstoreexpminusmax/gen/scalar-lut64-p2-x1.c.o
    #24 22.86 [540/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/f32-raddstoreexpminusmax/gen/scalar-lut64-p2-x4-acc2.c.o
    #24 22.87 [541/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/f32-raddstoreexpminusmax/gen/scalar-lut64-p2-x4.c.o
    #24 22.87 [542/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/f32-pavgpool/9p8x-minmax-scalar-c1.c.o
    #24 22.87 [543/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/f32-ppmm/gen/4x4-minmax-scalar.c.o
    #24 22.87 [544/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/f32-raddstoreexpminusmax/gen/scalar-lut64-p2-x2-acc2.c.o
    #24 22.87 [545/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/f32-raddstoreexpminusmax/gen/scalar-lut64-p2-x2.c.o
    #24 22.88 [546/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/f32-raddstoreexpminusmax/gen/scalar-lut64-p2-x4-acc4.c.o
    #24 22.88 [547/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/f32-raddstoreexpminusmax/gen/scalar-p5-x1.c.o
    #24 22.89 [548/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/f32-raddstoreexpminusmax/gen/scalar-p5-x2-acc2.c.o
    #24 22.90 [549/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/f32-raddstoreexpminusmax/gen/scalar-p5-x2.c.o
    #24 22.91 [550/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/f32-raddstoreexpminusmax/gen/scalar-p5-x4-acc4.c.o
    #24 22.91 [551/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/f32-raddstoreexpminusmax/gen/scalar-p5-x4.c.o
    #24 22.91 [552/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/f32-rmax/scalar.c.o
    #24 22.91 [553/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/f32-raddstoreexpminusmax/gen/scalar-p5-x4-acc2.c.o
    #24 22.91 [554/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/f32-relu/gen/scalar-x1.c.o
    #24 22.91 [555/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/f32-relu/gen/scalar-x2.c.o
    #24 22.91 [556/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/f32-relu/gen/scalar-x4.c.o
    #24 22.91 [557/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/f32-relu/gen/scalar-x8.c.o
    #24 22.92 [558/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/f32-sigmoid/gen/scalar-lut64-p2-div-x1.c.o
    #24 22.92 [559/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/f32-sigmoid/gen/scalar-lut2048-p1-div-x1.c.o
    #24 22.92 [560/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/f32-sigmoid/gen/scalar-p5-div-x1.c.o
    #24 22.92 [561/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/f32-spmm/gen/1x1-minmax-scalar-pipelined.c.o
    #24 22.93 [562/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/f32-spmm/gen/1x1-minmax-scalar.c.o
    #24 22.93 [563/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/f32-spmm/gen/2x1-minmax-scalar.c.o
    #24 22.94 [564/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/f32-sigmoid/gen/scalar-lut64-p2-div-x2.c.o
    #24 22.94 [565/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/f32-sigmoid/gen/scalar-lut2048-p1-div-x2.c.o
    #24 22.94 [566/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/f32-sigmoid/gen/scalar-p5-div-x2.c.o
    #24 22.94 [567/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/f32-spmm/gen/2x1-minmax-scalar-pipelined.c.o
    #24 22.94 [568/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/f32-spmm/gen/4x1-minmax-scalar.c.o
    #24 22.94 [569/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/f32-vbinary/gen/vadd-minmax-scalar-x1.c.o
    #24 22.94 [570/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/f32-vbinary/gen/vadd-minmax-scalar-x2.c.o
    #24 22.95 [571/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/f32-vbinary/gen/vadd-minmax-scalar-x4.c.o
    #24 22.95 [572/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/f32-vbinary/gen/vadd-relu-scalar-x1.c.o
    #24 22.95 [573/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/f32-vbinary/gen/vadd-relu-scalar-x2.c.o
    #24 22.95 [574/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/f32-sigmoid/gen/scalar-lut64-p2-div-x4.c.o
    #24 22.95 [575/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/f32-sigmoid/gen/scalar-lut2048-p1-div-x4.c.o
    #24 22.95 [576/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/f32-sigmoid/gen/scalar-p5-div-x4.c.o
    #24 22.95 [577/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/f32-spmm/gen/4x1-minmax-scalar-pipelined.c.o
    #24 22.96 [578/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/f32-vbinary/gen/vadd-minmax-scalar-x8.c.o
    #24 22.96 [579/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/f32-vbinary/gen/vadd-relu-scalar-x4.c.o
    #24 22.96 [580/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/f32-vbinary/gen/vadd-relu-scalar-x8.c.o
    #24 22.96 [581/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/f32-vbinary/gen/vadd-scalar-x2.c.o
    #24 22.96 [582/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/f32-vbinary/gen/vadd-scalar-x4.c.o
    #24 22.96 [583/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/f32-vbinary/gen/vadd-scalar-x8.c.o
    #24 22.96 [584/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/f32-vbinary/gen/vaddc-minmax-scalar-x1.c.o
    #24 22.96 [585/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/f32-spmm/gen/8x1-minmax-scalar-pipelined.c.o
    #24 22.97 [586/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/f32-vbinary/gen/vadd-scalar-x1.c.o
    #24 22.97 [587/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/f32-vbinary/gen/vaddc-minmax-scalar-x2.c.o
    #24 22.97 [588/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/f32-vbinary/gen/vaddc-minmax-scalar-x4.c.o
    #24 22.98 [589/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/f32-vbinary/gen/vaddc-relu-scalar-x1.c.o
    #24 22.98 [590/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/f32-vbinary/gen/vaddc-relu-scalar-x2.c.o
    #24 22.98 [591/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/f32-vbinary/gen/vaddc-relu-scalar-x4.c.o
    #24 22.98 [592/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/f32-spmm/gen/8x1-minmax-scalar.c.o
    #24 22.98 [593/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/f32-vbinary/gen/vaddc-minmax-scalar-x8.c.o
    #24 22.98 [594/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/f32-vbinary/gen/vaddc-relu-scalar-x8.c.o
    #24 22.98 [595/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/f32-vbinary/gen/vaddc-scalar-x1.c.o
    #24 22.98 [596/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/f32-vbinary/gen/vaddc-scalar-x2.c.o
    #24 22.99 [597/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/f32-vbinary/gen/vdiv-minmax-scalar-x2.c.o
    #24 22.99 [598/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/f32-spmm/gen/8x2-minmax-scalar.c.o
    #24 22.99 [599/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/f32-vbinary/gen/vdiv-relu-scalar-x1.c.o
    #24 22.99 [600/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/f32-vbinary/gen/vaddc-scalar-x4.c.o
    #24 22.99 [601/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/f32-vbinary/gen/vaddc-scalar-x8.c.o
    #24 23.00 [602/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/f32-vbinary/gen/vdiv-minmax-scalar-x1.c.o
    #24 23.00 [603/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/f32-vbinary/gen/vdiv-relu-scalar-x2.c.o
    #24 23.00 [604/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/f32-vbinary/gen/vdiv-minmax-scalar-x4.c.o
    #24 23.00 [605/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/f32-vbinary/gen/vdiv-relu-scalar-x8.c.o
    #24 23.01 [606/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/f32-vbinary/gen/vdiv-scalar-x1.c.o
    #24 23.01 [607/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/f32-vbinary/gen/vdiv-minmax-scalar-x8.c.o
    #24 23.03 [608/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/f32-vbinary/gen/vdivc-minmax-scalar-x2.c.o
    #24 23.03 [609/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/f32-vbinary/gen/vdiv-relu-scalar-x4.c.o
    #24 23.03 [610/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/f32-vbinary/gen/vdivc-minmax-scalar-x1.c.o
    #24 23.04 [611/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/f32-vbinary/gen/vdiv-scalar-x4.c.o
    #24 23.04 [612/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/f32-vbinary/gen/vdivc-scalar-x2.c.o
    #24 23.04 [613/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/f32-vbinary/gen/vdiv-scalar-x2.c.o
    #24 23.05 [614/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/f32-vbinary/gen/vdivc-minmax-scalar-x4.c.o
    #24 23.05 [615/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/f32-vbinary/gen/vdivc-relu-scalar-x1.c.o
    #24 23.05 [616/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/f32-vbinary/gen/vdivc-relu-scalar-x4.c.o
    #24 23.05 [617/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/f32-vbinary/gen/vmaxc-scalar-x1.c.o
    #24 23.05 [618/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/f32-vbinary/gen/vdiv-scalar-x8.c.o
    #24 23.05 [619/6115] Performing build step for 'nccl_external'
    #24 23.05 FAILED: nccl_external-prefix/src/nccl_external-stamp/nccl_external-build nccl/lib/libnccl_static.a 
    #24 23.05 cd /opt/pytorch/third_party/nccl/nccl && env make CXX=/usr/bin/c++ CUDA_HOME=/usr/local/cuda NVCC=/usr/local/cuda/bin/nvcc NVCC_GENCODE=-gencode=arch=compute_80,code=sm_80 BUILDDIR=/opt/pytorch/build/nccl VERBOSE=0 -j && /opt/conda/bin/cmake -E touch /opt/pytorch/build/nccl_external-prefix/src/nccl_external-stamp/nccl_external-build
    #24 23.05 make -C src build BUILDDIR=/opt/pytorch/build/nccl
    #24 23.05 make[1]: Entering directory '/opt/pytorch/third_party/nccl/nccl/src'
    #24 23.05 Grabbing   include/nccl_net.h                  > /opt/pytorch/build/nccl/include/nccl_net.h
    #24 23.05 Generating nccl.h.in                           > /opt/pytorch/build/nccl/include/nccl.h
    #24 23.05 Generating nccl.pc.in                          > /opt/pytorch/build/nccl/lib/pkgconfig/nccl.pc
    #24 23.05 Compiling  init.cc                             > /opt/pytorch/build/nccl/obj/init.o
    #24 23.05 Compiling  channel.cc                          > /opt/pytorch/build/nccl/obj/channel.o
    #24 23.05 Compiling  bootstrap.cc                        > /opt/pytorch/build/nccl/obj/bootstrap.o
    #24 23.05 Compiling  transport.cc                        > /opt/pytorch/build/nccl/obj/transport.o
    #24 23.05 Compiling  enqueue.cc                          > /opt/pytorch/build/nccl/obj/enqueue.o
    #24 23.05 Compiling  group.cc                            > /opt/pytorch/build/nccl/obj/group.o
    #24 23.05 Compiling  debug.cc                            > /opt/pytorch/build/nccl/obj/debug.o
    #24 23.05 Compiling  proxy.cc                            > /opt/pytorch/build/nccl/obj/proxy.o
    #24 23.05 Compiling  misc/nvmlwrap.cc                    > /opt/pytorch/build/nccl/obj/misc/nvmlwrap.o
    #24 23.05 Compiling  misc/ibvwrap.cc                     > /opt/pytorch/build/nccl/obj/misc/ibvwrap.o
    #24 23.05 Compiling  misc/utils.cc                       > /opt/pytorch/build/nccl/obj/misc/utils.o
    #24 23.05 Compiling  misc/argcheck.cc                    > /opt/pytorch/build/nccl/obj/misc/argcheck.o
    #24 23.05 Compiling  transport/p2p.cc                    > /opt/pytorch/build/nccl/obj/transport/p2p.o
    #24 23.05 Compiling  transport/shm.cc                    > /opt/pytorch/build/nccl/obj/transport/shm.o
    #24 23.05 Compiling  transport/net.cc                    > /opt/pytorch/build/nccl/obj/transport/net.o
    #24 23.05 Compiling  transport/net_socket.cc             > /opt/pytorch/build/nccl/obj/transport/net_socket.o
    #24 23.05 Compiling  transport/net_ib.cc                 > /opt/pytorch/build/nccl/obj/transport/net_ib.o
    #24 23.05 Compiling  transport/coll_net.cc               > /opt/pytorch/build/nccl/obj/transport/coll_net.o
    #24 23.05 Compiling  collectives/sendrecv.cc             > /opt/pytorch/build/nccl/obj/collectives/sendrecv.o
    #24 23.05 Compiling  collectives/all_reduce.cc           > /opt/pytorch/build/nccl/obj/collectives/all_reduce.o
    #24 23.05 Compiling  collectives/all_gather.cc           > /opt/pytorch/build/nccl/obj/collectives/all_gather.o
    #24 23.05 Compiling  collectives/broadcast.cc            > /opt/pytorch/build/nccl/obj/collectives/broadcast.o
    #24 23.05 Compiling  collectives/reduce.cc               > /opt/pytorch/build/nccl/obj/collectives/reduce.o
    #24 23.05 Compiling  collectives/reduce_scatter.cc       > /opt/pytorch/build/nccl/obj/collectives/reduce_scatter.o
    #24 23.05 Compiling  graph/topo.cc                       > /opt/pytorch/build/nccl/obj/graph/topo.o
    #24 23.05 Compiling  graph/paths.cc                      > /opt/pytorch/build/nccl/obj/graph/paths.o
    #24 23.05 Compiling  graph/search.cc                     > /opt/pytorch/build/nccl/obj/graph/search.o
    #24 23.05 Compiling  graph/connect.cc                    > /opt/pytorch/build/nccl/obj/graph/connect.o
    #24 23.05 Compiling  graph/rings.cc                      > /opt/pytorch/build/nccl/obj/graph/rings.o
    #24 23.05 Compiling  graph/trees.cc                      > /opt/pytorch/build/nccl/obj/graph/trees.o
    #24 23.05 Compiling  graph/tuning.cc                     > /opt/pytorch/build/nccl/obj/graph/tuning.o
    #24 23.05 Compiling  graph/xml.cc                        > /opt/pytorch/build/nccl/obj/graph/xml.o
    #24 23.05 make[2]: Entering directory '/opt/pytorch/third_party/nccl/nccl/src/collectives/device'
    #24 23.05 Generating rules                               > /opt/pytorch/build/nccl/obj/collectives/device/Makefile.rules
    #24 23.05 nvcc fatal   : Unsupported gpu architecture 'compute_80'
    #24 23.05 nvcc fatal   : Unsupported gpu architecture 'compute_80'
    #24 23.05 Makefile:52: recipe for target '/opt/pytorch/build/nccl/obj/collectives/device/sendrecv.dep' failed
    #24 23.05 make[2]: *** [/opt/pytorch/build/nccl/obj/collectives/device/sendrecv.dep] Error 1
    #24 23.05 make[2]: *** Waiting for unfinished jobs....
    #24 23.05 Makefile:52: recipe for target '/opt/pytorch/build/nccl/obj/collectives/device/all_reduce.dep' failed
    #24 23.05 make[2]: *** [/opt/pytorch/build/nccl/obj/collectives/device/all_reduce.dep] Error 1
    #24 23.05 nvcc fatal   : Unsupported gpu architecture 'compute_80'
    #24 23.05 Makefile:52: recipe for target '/opt/pytorch/build/nccl/obj/collectives/device/all_gather.dep' failed
    #24 23.05 make[2]: *** [/opt/pytorch/build/nccl/obj/collectives/device/all_gather.dep] Error 1
    #24 23.05 make[2]: *** wait: No child processes.  Stop.
    #24 23.05 Makefile:50: recipe for target '/opt/pytorch/build/nccl/obj/collectives/device/colldevice.a' failed
    #24 23.05 make[1]: *** [/opt/pytorch/build/nccl/obj/collectives/device/colldevice.a] Error 2
    #24 23.05 make[1]: *** Waiting for unfinished jobs....
    #24 23.05 make[1]: Leaving directory '/opt/pytorch/third_party/nccl/nccl/src'
    #24 23.05 Makefile:25: recipe for target 'src.build' failed
    #24 23.05 make: *** [src.build] Error 2
    #24 23.05 [620/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/f32-vbinary/gen/vdivc-minmax-scalar-x8.c.o
    #24 23.05 [621/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/f32-vbinary/gen/vdivc-scalar-x8.c.o
    #24 23.05 [622/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/f32-vbinary/gen/vdivc-relu-scalar-x2.c.o
    #24 23.06 [623/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/f32-vbinary/gen/vdivc-scalar-x1.c.o
    #24 23.06 [624/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/f32-vbinary/gen/vdivc-relu-scalar-x8.c.o
    #24 23.06 [625/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/f32-vbinary/gen/vmaxc-scalar-x8.c.o
    #24 23.06 [626/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/f32-vbinary/gen/vmul-minmax-scalar-x1.c.o
    #24 23.06 [627/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/f32-vbinary/gen/vmax-scalar-x2.c.o
    #24 23.06 [628/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/f32-vbinary/gen/vmul-minmax-scalar-x2.c.o
    #24 23.06 [629/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/f32-vbinary/gen/vmaxc-scalar-x2.c.o
    #24 23.06 [630/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/f32-vbinary/gen/vminc-scalar-x2.c.o
    #24 23.07 [631/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/f32-vbinary/gen/vmax-scalar-x1.c.o
    #24 23.07 [632/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/f32-vbinary/gen/vmin-scalar-x1.c.o
    #24 23.07 [633/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/f32-vbinary/gen/vmax-scalar-x4.c.o
    #24 23.07 [634/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/f32-vbinary/gen/vmul-minmax-scalar-x4.c.o
    #24 23.07 [635/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/f32-spmm/gen/8x4-minmax-scalar.c.o
    #24 23.07 [636/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/f32-vbinary/gen/vdivc-scalar-x4.c.o
    #24 23.08 [637/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/f32-vbinary/gen/vmaxc-scalar-x4.c.o
    #24 23.08 [638/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/f32-vbinary/gen/vmul-relu-scalar-x1.c.o
    #24 23.08 [639/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/f32-vbinary/gen/vmax-scalar-x8.c.o
    #24 23.08 [640/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/f32-vbinary/gen/vmin-scalar-x2.c.o
    #24 23.08 [641/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/f32-vbinary/gen/vmul-relu-scalar-x4.c.o
    #24 23.08 [642/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/f32-vbinary/gen/vmin-scalar-x4.c.o
    #24 23.08 [643/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/f32-vbinary/gen/vminc-scalar-x8.c.o
    #24 23.08 [644/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/f32-vbinary/gen/vminc-scalar-x1.c.o
    #24 23.09 [645/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/f32-vbinary/gen/vmul-scalar-x8.c.o
    #24 23.09 [646/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/f32-vbinary/gen/vmulc-minmax-scalar-x4.c.o
    #24 23.09 [647/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/f32-vbinary/gen/vmul-scalar-x4.c.o
    #24 23.09 [648/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/f32-vbinary/gen/vmul-scalar-x1.c.o
    #24 23.09 [649/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/f32-vbinary/gen/vmulc-minmax-scalar-x2.c.o
    #24 23.09 [650/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/f32-vbinary/gen/vmulc-relu-scalar-x1.c.o
    #24 23.10 [651/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/f32-vbinary/gen/vmul-relu-scalar-x2.c.o
    #24 23.10 [652/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/f32-vbinary/gen/vmulc-minmax-scalar-x1.c.o
    #24 23.10 [653/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/f32-vbinary/gen/vmul-scalar-x2.c.o
    #24 23.10 [654/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/f32-vbinary/gen/vmulc-minmax-scalar-x8.c.o
    #24 23.10 [655/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/f32-vbinary/gen/vminc-scalar-x4.c.o
    #24 23.10 [656/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/f32-vbinary/gen/vmin-scalar-x8.c.o
    #24 23.10 [657/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/f32-vbinary/gen/vmul-relu-scalar-x8.c.o
    #24 23.10 [658/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/f32-vbinary/gen/vmulc-relu-scalar-x2.c.o
    #24 23.11 [659/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/f32-vbinary/gen/vmul-minmax-scalar-x8.c.o
    #24 23.11 [660/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/f32-vbinary/gen/vmulc-relu-scalar-x4.c.o
    #24 23.11 [661/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/f32-vbinary/gen/vmulc-relu-scalar-x8.c.o
    #24 23.11 [662/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/f32-vbinary/gen/vrdivc-minmax-scalar-x1.c.o
    #24 23.12 [663/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/f32-vbinary/gen/vmulc-scalar-x1.c.o
    #24 23.12 [664/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/f32-vbinary/gen/vmulc-scalar-x8.c.o
    #24 23.12 [665/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/f32-vbinary/gen/vmulc-scalar-x4.c.o
    #24 23.12 [666/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/f32-vbinary/gen/vmulc-scalar-x2.c.o
    #24 23.12 [667/6115] Building CXX object third_party/protobuf/cmake/CMakeFiles/libprotobuf.dir/__/src/google/protobuf/struct.pb.cc.o
    #24 23.12 [668/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/f32-vbinary/gen/vrdivc-minmax-scalar-x2.c.o
    #24 23.13 [669/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/f32-vbinary/gen/vrdivc-minmax-scalar-x4.c.o
    #24 23.13 [670/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/f32-vbinary/gen/vrdivc-minmax-scalar-x8.c.o
    #24 23.13 [671/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/f32-dwconv2d-chw/gen/3x3p1-minmax-sse-6x4.c.o
    #24 23.15 [672/6115] Building C object confu-deps/XNNPACK/CMakeFiles/XNNPACK.dir/src/packing.c.o
    #24 23.20 [673/6115] Building CXX object third_party/protobuf/cmake/CMakeFiles/libprotobuf.dir/__/src/google/protobuf/wire_format.cc.o
    #24 23.21 [674/6115] Building CXX object third_party/protobuf/cmake/CMakeFiles/libprotobuf.dir/__/src/google/protobuf/generated_message_table_driven.cc.o
    #24 23.21 [675/6115] Building CXX object third_party/protobuf/cmake/CMakeFiles/libprotobuf.dir/__/src/google/protobuf/descriptor_database.cc.o
    #24 23.37 [676/6115] Building CXX object third_party/protobuf/cmake/CMakeFiles/libprotobuf.dir/__/src/google/protobuf/compiler/parser.cc.o
    #24 23.51 [677/6115] Building CXX object third_party/protobuf/cmake/CMakeFiles/libprotobuf.dir/__/src/google/protobuf/generated_message_reflection.cc.o
    #24 23.70 [678/6115] Building CXX object third_party/protobuf/cmake/CMakeFiles/libprotobuf.dir/__/src/google/protobuf/util/message_differencer.cc.o
    #24 23.90 [679/6115] Building CXX object third_party/protobuf/cmake/CMakeFiles/libprotobuf.dir/__/src/google/protobuf/text_format.cc.o
    #24 24.51 [680/6115] Generating src/x86_64-fma/2d-fourier-8x8.py.o
    #24 25.33 [681/6115] Building CXX object third_party/protobuf/cmake/CMakeFiles/libprotobuf.dir/__/src/google/protobuf/descriptor.pb.cc.o
    #24 28.10 [682/6115] Building CXX object third_party/protobuf/cmake/CMakeFiles/libprotobuf.dir/__/src/google/protobuf/descriptor.cc.o
    #24 28.10 ninja: build stopped: subcommand failed.
    ------
    executor failed running [/bin/sh -c USE_CUDA=1 USE_CUDNN=1     TORCH_NVCC_FLAGS=${TORCH_NVCC_FLAGS}     TORCH_CUDA_ARCH_LIST=${TORCH_CUDA_ARCH_LIST}     CMAKE_PREFIX_PATH="$(dirname $(which conda))/../"     python setup.py bdist_wheel -d /tmp/dist]: exit code: 1
    make: *** [Makefile:104: build-torch-full] Error 1
    

    Is this normal? what should I do? Thank you in advance!

    opened by pvnieo 11
  • GPUs Not Detected (RTX 3090; Torch 1.9.1)

    GPUs Not Detected (RTX 3090; Torch 1.9.1)

    Hello, and thank you for putting together this resource!

    I tried following the steps in the README without deviation, but using the resulting docker container doesn't detect the GPUs in a Python context. I'm using a system with 3x RTX 3090s

    Steps to Reproduce

    1. Clone this directory
    2. Run the following command (taken straight from the README)
    make all CC="8.6" PYTORCH_VERSION_TAG=v1.9.1 TORCHVISION_VERSION_TAG=v0.10.1 TORCHTEXT_VERSION_TAG=v0.10.1 TORCHAUDIO_VERSION_TAG=v0.9.1
    
    1. Install docker-compose v2
    2. Edit docker-compose.yaml Before Editing: https://github.com/veritas9872/PyTorch-Universal-Docker-Template/blob/f8708d44c9d3a013155cd91985534da1a4eee7fb/docker-compose.yaml#L28 After Editing:
            TORCH_CUDA_ARCH_LIST: ${CC:-'8.6+PTX'}
    

    PS - also edited the GID and UID as suggested but not posting here as that's system specific 5. Run the following commands (taken from the README):

    docker compose up -d train
    docker compose exec train /bin/bash
    
    1. Open up ipython and run the following
    import torch
    torch.cuda.is_available()
    

    to get this error:

    /opt/conda/lib/python3.8/site-packages/torch/cuda/__init__.py:52: UserWarning: CUDA initialization: Unexpected error from cudaGetDeviceCount(). Did you run some cuda functions before calling NumCudaDevices() that might have already set an error? Error 804: forward compatibility was attempted on non supported HW (Triggered internally at  ../c10/cuda/CUDAFunctions.cpp:115.)
      return torch._C._cuda_getDeviceCount() > 0
    

    Thank you!

    opened by rsomani95 10
  • git clone --job 에러

    git clone --job 에러

    git clone 명령어에 대하여 unknown option `jobs' 에러가 발생합니다. 혹시 해결방법이 있을까요?


    [build-install 3/6] RUN git clone --recursive --jobs 0 https://github.com/pytorch/pytorch.git:
    #16 0.633 error: unknown option `jobs'
    #16 0.633 usage: git clone [] [--] [

    ]
    #16 0.633 #16 0.633 -v, --verbose be more verbose #16 0.633 -q, --quiet be more quiet #16 0.633 --progress force progress reporting #16 0.633 -n, --no-checkout don't create a checkout #16 0.633 --bare create a bare repository #16 0.633 --mirror create a mirror repository (implies bare) #16 0.633 -l, --local to clone from a local repository #16 0.633 --no-hardlinks don't use local hardlinks, always copy #16 0.633 -s, --shared setup as shared repository #16 0.633 --recursive initialize submodules in the clone #16 0.633 --recurse-submodules initialize submodules in the clone #16 0.633 --template #16 0.633 directory from which templates will be used #16 0.633 --reference reference repository #16 0.633 --dissociate use --reference only while cloning #16 0.633 -o, --origin use instead of 'origin' to track upstream #16 0.633 -b, --branch #16 0.633 checkout instead of the remote's HEAD #16 0.633 -u, --upload-pack #16 0.633 path to git-upload-pack on the remote #16 0.633 --depth create a shallow clone of that depth #16 0.633 --single-branch clone only one branch, HEAD or --branch #16 0.633 --separate-git-dir #16 0.633 separate git dir from working tree #16 0.633 -c, --config <key=value> #16 0.633 set config inside the new repository #16 0.633


    executor failed running [/bin/sh -c git clone --recursive --jobs 0 https://github.com/pytorch/pytorch.git]: exit code: 129 Makefile:23: recipe for target 'build-install' failed make: *** [build-install] Error 1

    opened by kboseong 8
  • A minimal way to adopt the method in an existing project?

    A minimal way to adopt the method in an existing project?

    Hello,

    I came across this template when having trouble installing the environment of an existing project, given as a conda environment, but without more details about system dependencies. I wanted to build a portable environment with Docker but existing PyTorch images do not support my old GPU kernel, so I was looking for a way to compile Pytorch and Torchvision (CUDA enabled) from source in a Docker environment.

    This project template addresses my problem and doubles down with best practices, however, I found it overwhelming to use/adapt it to my current project. Mainly because of the high complexity of building the environment that is transferred to the user (E.g. setting up the different env variables).

    Is there a minimalist way to take this template and integrate it in an existing project?

    Thanks!

    opened by skandermoalla 7
  • Image size is 756GB.

    Image size is 756GB.

    The image size is 756GB with the command "make all-full CC="7.5" on the server with 40 cores and 8 Gtx2080-Ti s.

    I am not sure if the image is taking really this big size, but when I tried to push it to Docker Hub, it really went 756GB. (I am pushing it now, currently, 41.48GB/756GB pushed)

    opened by WantToLearnJapanese 7
  • Torchvision does not build PYTORCH_VERSION_TAG:-v1.10.2 TORCHVISION_VERSION_TAG:-v0.11.3

    Torchvision does not build PYTORCH_VERSION_TAG:-v1.10.2 TORCHVISION_VERSION_TAG:-v0.11.3

    Hello,

    I'm building the docker image with the following config (all others being default)

    BUILD_MODE=include
    CCA=3.5
    CUDA_VERSION:-11.3.1
    PYTHON_VERSION:-3.8
    PYTORCH_VERSION_TAG:-v1.10.2
    TORCHVISION_VERSION_TAG:-v0.11.3
    

    The build fails at the build-vision stage with the following error:

    #0 85.06 [14/16] c++ -MMD -MF /opt/vision/build/temp.linux-x86_64-3.8/opt/vision/torchvision/csrc/io/video_reader/video_reader.o.d -pthread -B /opt/conda/compiler_compat -Wno-unused-result -Wsign-compare -DNDEBUG -fwrapv -O2 -Wall
    -fPIC -O2 -isystem /opt/conda/include -fPIC -O2 -isystem /opt/conda/include -fPIC -I/opt/vision/torchvision/csrc/io/decoder -I/opt/vision/torchvision/csrc/io/video_reader -I/opt/vision/torchvision/csrc/io/video -I/opt/vision/torchv
    ision/csrc -I/opt/conda/include -I/opt/conda/include/x86_64-linux-gnu -I/opt/vision/torchvision/csrc -I/opt/conda/lib/python3.8/site-packages/torch/include -I/opt/conda/lib/python3.8/site-packages/torch/include/torch/csrc/api/inclu
    de -I/opt/conda/lib/python3.8/site-packages/torch/include/TH -I/opt/conda/lib/python3.8/site-packages/torch/include/THC -I/usr/local/cuda/include -I/opt/conda/lib/python3.8/site-packages/torch/include -I/opt/conda/lib/python3.8/sit
    e-packages/torch/include/torch/csrc/api/include -I/opt/conda/lib/python3.8/site-packages/torch/include/TH -I/opt/conda/lib/python3.8/site-packages/torch/include/THC -I/opt/conda/include/python3.8 -c -c /opt/vision/torchvision/csrc/
    io/video_reader/video_reader.cpp -o /opt/vision/build/temp.linux-x86_64-3.8/opt/vision/torchvision/csrc/io/video_reader/video_reader.o -std=c++14 -DTORCH_API_INCLUDE_EXTENSION_H '-DPYBIND11_COMPILER_TYPE="_gcc"' '-DPYBIND11_STDLIB=
    "_libstdcpp"' '-DPYBIND11_BUILD_ABI="_cxxabi1013"' -DTORCH_EXTENSION_NAME=video_reader -D_GLIBCXX_USE_CXX11_ABI=1
    #0 85.06 ninja: build stopped: subcommand failed.
    #0 85.06 Traceback (most recent call last):
    #0 85.06   File "/opt/conda/lib/python3.8/site-packages/torch/utils/cpp_extension.py", line 1717, in _run_ninja_build
    #0 85.06     subprocess.run(
    #0 85.06   File "/opt/conda/lib/python3.8/subprocess.py", line 516, in run
    #0 85.06     raise CalledProcessError(retcode, process.args,
    #0 85.06 subprocess.CalledProcessError: Command '['ninja', '-v']' returned non-zero exit status 1.
    #0 85.06
    #0 85.06 The above exception was the direct cause of the following exception:                                                    
    #0 85.06
    #0 85.06 Traceback (most recent call last):
    #0 85.06   File "setup.py", line 484, in <module>
    #0 85.06     setup(
    #0 85.06   File "/opt/conda/lib/python3.8/site-packages/setuptools/__init__.py", line 153, in setup
    #0 85.06     return distutils.core.setup(**attrs)
    #0 85.06   File "/opt/conda/lib/python3.8/distutils/core.py", line 148, in setup
    #0 85.06     dist.run_commands()
    #0 85.06   File "/opt/conda/lib/python3.8/distutils/dist.py", line 966, in run_commands
    #0 85.06     self.run_command(cmd)
    #0 85.06   File "/opt/conda/lib/python3.8/distutils/dist.py", line 985, in run_command
    #0 85.06     cmd_obj.run()
    #0 85.06   File "/opt/conda/lib/python3.8/site-packages/wheel/bdist_wheel.py", line 299, in run
    #0 85.06     self.run_command('build')
    #0 85.06   File "/opt/conda/lib/python3.8/distutils/cmd.py", line 313, in run_command
    #0 85.06     self.distribution.run_command(command)
    #0 85.06   File "/opt/conda/lib/python3.8/distutils/dist.py", line 985, in run_command
    #0 85.06     cmd_obj.run()
    #0 85.06   File "/opt/conda/lib/python3.8/distutils/command/build.py", line 135, in run
    #0 85.06     self.run_command(cmd_name)
    #0 85.06   File "/opt/conda/lib/python3.8/distutils/cmd.py", line 313, in run_command
    #0 85.06     self.distribution.run_command(command)
    #0 85.06   File "/opt/conda/lib/python3.8/distutils/dist.py", line 985, in run_command
    #0 85.06     cmd_obj.run()
    #0 85.06   File "/opt/conda/lib/python3.8/site-packages/setuptools/command/build_ext.py", line 79, in run
    #0 85.07     _build_ext.run(self)
    #0 85.07   File "/opt/conda/lib/python3.8/distutils/command/build_ext.py", line 340, in run
    #0 85.07     self.build_extensions()
    #0 85.07   File "/opt/conda/lib/python3.8/site-packages/torch/utils/cpp_extension.py", line 735, in build_extensions
    #0 85.07     build_ext.build_extensions(self)
    #0 85.07   File "/opt/conda/lib/python3.8/distutils/command/build_ext.py", line 449, in build_extensions
    #0 85.07     self._build_extensions_serial()
    #0 85.07   File "/opt/conda/lib/python3.8/distutils/command/build_ext.py", line 474, in _build_extensions_serial
    #0 85.07     self.build_extension(ext)
    #0 85.07   File "/opt/conda/lib/python3.8/site-packages/setuptools/command/build_ext.py", line 202, in build_extension
    #0 85.07     _build_ext.build_extension(self, ext)
    #0 85.07   File "/opt/conda/lib/python3.8/distutils/command/build_ext.py", line 528, in build_extension
    #0 85.07     objects = self.compiler.compile(sources,
    #0 85.07   File "/opt/conda/lib/python3.8/site-packages/torch/utils/cpp_extension.py", line 556, in unix_wrap_ninja_compile
    #0 85.07     _write_ninja_file_and_compile_objects(
    #0 85.07   File "/opt/conda/lib/python3.8/site-packages/torch/utils/cpp_extension.py", line 1399, in _write_ninja_file_and_compile_objects
    #0 85.07     _run_ninja_build(
    #0 85.07   File "/opt/conda/lib/python3.8/site-packages/torch/utils/cpp_extension.py", line 1733, in _run_ninja_build
    #0 85.07     raise RuntimeError(message) from e
    #0 85.07 RuntimeError: Error compiling objects for extension
    ------
    failed to solve: executor failed running [/bin/sh -c python setup.py bdist_wheel -d /tmp/dist]: exit code: 1
    make: *** [Makefile:67: build] Error 17
    

    When ignoring the build-vision stage, the build works fine. I.e by changing the train-builds-include as

    FROM ${BUILD_IMAGE} AS train-builds-include
    
    ...
    
    COPY --link --from=install-base /opt/conda /opt/conda
    COPY --link --from=build-pillow /tmp/dist  /tmp/dist
    
    COPY --link --from=build-torch /tmp/dist  /tmp/dist <--- here 
    
    COPY --link --from=fetch-pure   /opt/zsh   /opt
    

    I get the same error with different versions of torchvision. I tried 0.11.3, 0.11.2 and 0.11.1.

    Could you suggest any help?

    Thanks a lot!

    opened by skandermoalla 5
  • Conda dependencies in `train` stage

    Conda dependencies in `train` stage

    Hello,

    At the moment the template supports installing dependencies withpip in the train stage specified in pip-train.requirements.txt. Is there a way to add conda dependencies as well? Best practices are to install the conda dependencies first then the pip ones.

    Thanks!

    opened by skandermoalla 5
  • Building the template for x86_64 or VM

    Building the template for x86_64 or VM

    Hi

    Thanks for the detailed steps on building and documentation. Can I make the build image for intel desktop or VM ? In case I use (build/train) on a VM what are the TARGET_ARCH options that needs to be given ?

    Thank you so much Srini

    opened by skanduru 5
  • Building from source having x4 speedup vs. naive install

    Building from source having x4 speedup vs. naive install

    Hi. Thank you for releasing this library.

    I had a question about this tidbit in the README:

    PyTorch built from source can be x4 faster than a naïve PyTorch install.

    Is there any past discussion or justification about this speedup? Is source install faster both on CPU and GPU? Could it be attributed to better optimization of BLAS or MKL?

    Would love to know more about this.

    opened by seongminp 5
  • Fix PyTorch download URL variables

    Fix PyTorch download URL variables

    Fix inconsistencies in the PyTorch download URL variables:

    • Dockerfile defined PYTORCH_INDEX_URL
    • but compose file defines PYTORCH_URL (here)

    Also, delete default values in Dockerfile as the compose file specifies them.

    opened by skandermoalla 4
  • make build failed

    make build failed

    .env is below:

    CUDA_VERSION=10.2
    CUDNN_VERSION=8
    PYTHON_VERSION=3.7
    TORCH_CUDA_ARCH_LIST=8.0
    PYTORCH_VERSION_TAG=v1.11.0
    

    make build met error:

     > [internal] load metadata for docker.io/nvidia/cuda:11.2-cudnn8-devel-ubuntu18.04:
    ------
    failed to solve: rpc error: code = Unknown desc = failed to solve with frontend dockerfile.v0: failed to solve with frontend gateway.v0: rpc error: code = Unknown desc = docker.io/nvidia/cuda:11.2-cudnn8-devel-ubuntu18.04: not found
    Makefile:55: recipe for target 'build' failed
    

    seems nvidia changes image names?

    opened by Jack47 4
Releases(v0.5.0)
  • v0.5.0(Jan 4, 2023)

    Changed the training environment so that conda is now the preferred package manager instead of just the virtual environment. Fixed many bugs and issues to make the project easier to use.

    Source code(tar.gz)
    Source code(zip)
  • v0.4.1(Sep 12, 2022)

    Separate out installing PyTorch and related libraries. Rename download-only stages as fetch for better contrast with build stages. Updates in the documentation.

    Source code(tar.gz)
    Source code(zip)
  • v0.4.0(Aug 30, 2022)

  • v0.3.0(Jun 1, 2022)

    Support for external files for installation. The NSight systems debian binary cannot be installed via the command line. An external .deb file is therefore used. Also, Git Large File System (LFS) is used to prevent the .git directory from bloating. The build guidelines have also changed in order to make the meaning clearer (I think). CCA is no longer mandatory and the .env file is guarded by a separate recipe. Python package installation is now fully parallel with apt package installation. Finally figured out how to get separate volume paths for different hosts. Using docker-compose.override.yaml was decided to be the best approach. Documentation still needs work.

    Source code(tar.gz)
    Source code(zip)
  • v0.2.2(Apr 30, 2022)

  • v0.2.1(Apr 3, 2022)

    This is a hack release forced by the sudden failure of URLs in TorchAudio and the Kakao mirror for PyPI. Although both have simple fixes, these issues mean that users must become more involved with the details. To ensure that the build does not fail for new users, TorchAudio source builds and using PyPI mirrors have been temporarily disabled. These functionalities will be restored ASAP.

    Source code(tar.gz)
    Source code(zip)
  • v0.2.0(Mar 26, 2022)

    Summary

    Total makeover of the entire project from the ground up. Most documentation concerning how to use the Makefile for building Docker images and speed gains have been removed. More documentation and examples will come soon.

    What's Changed

    • Hotfix password by @veritas9872 in https://github.com/cresset-template/cresset/pull/24
    • Updates for Cresset by @veritas9872 in https://github.com/cresset-template/cresset/pull/25
    • Version 0.2 breaking update. by @veritas9872 in https://github.com/cresset-template/cresset/pull/27

    Full Changelog: https://github.com/cresset-template/cresset/compare/v0.1.2...v0.2.0

    Source code(tar.gz)
    Source code(zip)
  • v0.2.0-rc2(Jan 28, 2022)

    What's Changed

    • Hotfix password by @veritas9872 in https://github.com/cresset-template/cresset/pull/24
    • Updates for Cresset by @veritas9872 in https://github.com/cresset-template/cresset/pull/25
    • Version 0.2 breaking update. by @veritas9872 in https://github.com/cresset-template/cresset/pull/27

    Full Changelog: https://github.com/cresset-template/cresset/compare/v0.1.2...v0.2.0-rc2

    Source code(tar.gz)
    Source code(zip)
  • v0.2.0-rc1(Jan 23, 2022)

    New Features:

    1. Ubuntu 16.04 LTS support (not tested rigorously).
    2. CentOS and UBI support (not tested at all).
    3. MKL removal support for non-intel CPUs.
    4. Implementation of apt and pip package requirements files.
    5. Build-time optimizations for much faster build speeds. Caffe2 is disabled by default. Build tests are also disabled by default.
    6. LLD v11.2 used for faster linking during compilation.
    7. Jemalloc 5.2.1 used as the memory allocator for faster build and runtimes.
    8. Utilization of conda-forge and intel channels for optimized libraries.
    9. Addition of new docker-compose services for development.
    10. Inclusion of missing variable guards for the compute capability variable CCA during initialization of Docker Compose and Makefile build.

    Breaking Changes:

    1. The Compute Capability variable CC has been renamed CCA to avoid clashing with the C Compiler CC variable.
    2. Project renamed Cresset for greater memorability.

    Also fixed many bugs that I found.

    Source code(tar.gz)
    Source code(zip)
  • v0.1.2(Dec 30, 2021)

    Added new zsh functionality such as auto-complete and syntax highlighting. If these distract more than they help, simply comment out the installation. The NGC image had a bug where shell was fixed to bash and could not be changed to sh. This was (sort-of) fixed by using printf instead of echo. Also, the NGC image was not using the mirror link due to an additional pip.conf file inside /opt/conda. The code was accommodated to manage. Finally, the docker-compose.yaml full service had a bug in listing image to cache from. This too was fixed.

    Source code(tar.gz)
    Source code(zip)
  • v0.1.1(Dec 27, 2021)

    Mostly stylistic updates and some minor bugfixes. Benchmarking code is now cleaner, with CPU benchmarking also available. Explanation of how to use the .env file in make commands now included.

    Source code(tar.gz)
    Source code(zip)
  • v0.1.0(Dec 26, 2021)

    Initial Stable Release

    First release with semantic versioning.

    The project is now rather stable and thoroughly debugged, seeing as how there have been no bug reports for over a month.

    I have decided to adopt semantic versioning, even though I am worried that some people will not use the project now that there is a scary v0.x in front. Rest assured that the project works very well and that bugs are most likely due to the host environment or something the user did, not the project code itself.

    What's changed

    Unlike the previous versions (which I will call the alpha version from now on), the default shell is now zsh instead of bash. I have set the pure prompt as the default because of its compatibility with most terminals.

    The project can now use requirements.txt files, which was a feature in very high demand.

    I have also included benchmark.py for easy benchmarking. This was the most highly sought after functionality due to my claims of speed enhancements. Now anyone can measure the speed differences for themselves.

    Slightly friendlier README organization. No images though.

    Source code(tar.gz)
    Source code(zip)
Owner
Joonhyung Lee/이준형
Sapere Aude! Dare to know. Dare to understand.
Joonhyung Lee/이준형
Template for creating ds simple projects

ds-project-template Template for creating ds simple projects Requirements pyenv python==3.9.4 Setup For this purpose you use following commands: pytho

null 1 Dec 17, 2021
Cookiecutter Flask OpenAPI is a template for jumpstarting production-ready Flask projects quickly.

Cookiecutter Flask OpenAPI is a template for jumpstarting production-ready Flask projects quickly. It has a well organized and scalable structure. It uses API design first

Roger Camargo 3 Nov 17, 2022
A low dependency and really simple to start project template for Python Projects.

Python Project Template A low dependency and really simple to start project template for Python Projects. HOW TO USE THIS TEMPLATE DO NOT FORK this is

Yurii Dubinka 5 Jan 21, 2022
Forkable, Minimal Template for Starknet Projects.

Forkable, Minimal Template for Starknet Projects.

velleity.eth 44 Oct 9, 2022
Cookiecutter-allpurpose-minimal-python - A simple cookiecutter template for general-purpose python projects.

cookiecutter-allpurpose-minimal-python A simple cookiecutter template for general-purpose python projects. To use, run pip install cookiecutter cookie

E. Tolga Ayan 2 Jan 24, 2022
Ultimate Django3.2 Template for starting any project from not zero!

Ultimate Django3.2 Template for starting any project from not zero!

TheAliBigdeli 37 Dec 20, 2022
A template repository implementing HTML5 Boilerplate 8.0 in Sanic using the Domonic framework.

sanic-domonic-h5bp A template repository implementing HTML5 Boilerplate 8.0 in Sanic using the Domonic framework. If you need frontend interactivity,

PyXY 3 Dec 12, 2022
Bleeding edge django template focused on code quality and security.

wemake-django-template Bleeding edge django2.2 template focused on code quality and security. Purpose This project is used to scaffold a django projec

wemake.services 1.6k Jan 4, 2023
Project template layout for Django 3.0+

Django 3.0+ project template This is a simple Django 3.0+ project template with my preferred setup. Most Django project templates make way too many as

José Padilla 649 Dec 30, 2022
Bleeding edge django template focused on code quality and security.

wemake-django-template Bleeding edge django2.2 template focused on code quality and security. Purpose This project is used to scaffold a django projec

wemake.services 1.6k Jan 8, 2023
Django Webpack starter template for using Webpack 5 with Django 3.1 & Bootstrap 4. Yes, it can hot-reload.

Django Webpack Starter Hello fellow human. The repo uses Python 3.9.* Django 3.1.* Webpack 5.4.* Bootstrap 4.5.* Pipenv If you have any questions twe

Ganesh Khade 56 Nov 28, 2022
A Django starter template with a sound foundation.

SOS Django Template SOS Django Tempalate is a Django starter template that has opinionated and good solutions while starting your new Django project.

Eray Erdin 19 Oct 30, 2022
Generic template for python service

Cookie cutter template example Technology stack Flask Gevent UWSGI Poetry Docker Docker-compose Installation pip install cookiecutter cookiecutter git

Churkin Oleg 11 Oct 22, 2022
A template for some new Python tool or package with a reasonable basic setup.

python-app-template A template with a reasonable basic setup, including: black (formatting) flake8 (linting) mypy (type checking) isort (import sortin

Anton Pirogov 3 Jul 19, 2022
Generic python project template

generic-python-project-template generic-python-project-template STEPS - STEP 01- Create a repository by using template repository STEP 02- Clone the n

SUNNY BHAVEEN CHANDRA 3 Oct 3, 2022
Basic Docker Compose template application with Flask, Celery, Redis, MySQL, SocketIO, Nginx and Gunicorn.

Nginx / Gunicorn / Flask ?? / Celery / SocketIO / MySQL / Redis / Docker ?? sample application Basic Docker Compose template application for orchestat

Alex Oarga 8 Aug 6, 2022
Code Kata Python Template

Code Kata Python Template This is the code kata template for python created by Aula de Software Libre de la Universidad de Córdoba Step 1. Use this re

Sergio Gómez 2 Nov 30, 2021
A python starter package to be used as a template for creating your own python packages.

Python Starter Package This is a basic python starter package to be used as a template for creating your own python packages. Github repo: https://git

Mystic 1 Apr 4, 2022
Template to quickly start your playwright-python project

Playwright-python template ?? Template to quickly start your playwright-python project Getting started • Demo • Configuration Getting started Clone th

Constantin 1 Dec 13, 2021