Update Dockerfile.base.py36-cuda11.2

Author: hephaex

commons/Dockerfile.base.py36-cuda11.2

@@ -5,8 +5,7 @@ ted carefully
 ARG CUDA=11.2
 ARG LIBNVINFER=8.0.0-1
 ARG LIBNVINFER_MAJOR_VERSION=8
-#ARG LIBNVINFER=7.2.3-1
-#ARG LIBNVINFER_MAJOR_VERSION=7
+ENV MLNX_OFED_VERSION=5.4-3.1.0.0
 ARG CUDNN=8.0.5.39-1
 ENV NCCL=2.8.4
 ENV NCCL_VERSION=2.8.4-1+cuda11.2
@@ -27,36 +26,27 @@ ENV DEBIAN_FRONTEND=noninteractive \
     PYTHONUNBUFFERED=1 \
     LIBRARY_PATH=/usr/local/cuda/lib64/stubs \
     _CUDA_COMPAT_PATH="/usr/local/cuda/compat" \
-    LD_LIBRARY_PATH="/usr/local/cuda/extras/CUPTI/lib64:/usr/local/cuda/compat/\
-lib:/usr/local/nvidia/lib:/usr/local/nvidia/lib64:/usr/local/cuda/lib64:/usr/in\
-clude/x86_64-linux-gnu:/opt/intel/compilers_and_libraries_2020.3.279/linux/mkl/\
-lib/intel64" \
-    PATH="/usr/local/mpi/bin:/usr/local/ucx:/usr/local/nvidia/bin:/usr/local/cu\
-da/bin:/usr/local/sbin:/usr/bin/cmake/bin:/usr/local/bin:/usr/sbin:/usr/bin:/sb\
-in:/bin:/opt/tensorrt/bin:/usr/local/src/lightgbm/LightGBM:/usr/local/bin/mecab\
-" \
+    LD_LIBRARY_PATH="/usr/local/cuda/extras/CUPTI/lib64:/usr/local/cuda/compat/lib:/usr/local/nvidia/lib:/usr/local/nvidia/lib64:/usr/local/cuda/lib64:/usr/in\
+clude/x86_64-linux-gnu:/opt/intel/compilers_and_libraries_2020.3.279/linux/mkl/lib/intel64" \
+    PATH="/usr/local/mpi/bin:/usr/local/ucx:/usr/local/nvidia/bin:/usr/local/cuda/bin:/usr/local/sbin:/usr/bin/cmake/bin:/usr/local/bin:/usr/sbin:/usr/bin:/sb\
+in:/bin:/opt/tensorrt/bin:/usr/local/src/lightgbm/LightGBM:/usr/local/bin/mecab" \
     LANG=C.UTF-8
 
 # Set default shell to /bin/bash
 SHELL ["/bin/bash", "-cu"]
 
 RUN rm -fr /var/lib/apt/lists/* && \
     apt-get update -y && \
-    apt-get install -y --no-install-recommends software-properties-common wget \
-curl && \
+    apt-get install -y --no-install-recommends software-properties-common wget curl && \
     curl -sL https://deb.nodesource.com/setup_12.x | bash - && \
-    add-apt-repository -y "deb http://security.ubuntu.com/ubuntu xenial-securit\
-y main" && \
+    add-apt-repository -y "deb http://security.ubuntu.com/ubuntu xenial-security main" && \
     add-apt-repository -y ppa:deadsnakes/ppa && \
     add-apt-repository -y ppa:bashtop-monitor/bashtop && \
-    apt-key adv --fetch-keys https://developer.download.nvidia.com/compute/cuda\
-/repos/ubuntu2004/x86_64/7fa2af80.pub &&\
-    wget https://apt.repos.intel.com/intel-gpg-keys/GPG-PUB-KEY-INTEL-SW-PRODUC\
-TS-2019.PUB && \
+    apt-key adv --fetch-keys https://developer.download.nvidia.com/compute/cuda/repos/ubuntu2004/x86_64/7fa2af80.pub &&\
+    wget https://apt.repos.intel.com/intel-gpg-keys/GPG-PUB-KEY-INTEL-SW-PRODUCTS-2019.PUB && \
     apt-key add GPG-PUB-KEY-INTEL-SW-PRODUCTS-2019.PUB && \
     rm GPG-PUB-KEY-INTEL-SW-PRODUCTS-2019.PUB && \
-    wget https://apt.repos.intel.com/setup/intelproducts.list -O /etc/apt/sourc\
-es.list.d/intelproducts.list && \
+    wget https://apt.repos.intel.com/setup/intelproducts.list -O /etc/apt/sources.list.d/intelproducts.list && \
     apt-get update -y && \
     apt-get install -y --no-install-recommends \
         autoconf \
@@ -67,7 +57,7 @@ es.list.d/intelproducts.list && \
         curl \
         devscripts \
         dh-make \
-                dkms \
+        dkms \
         dkms \
         fakeroot \
         ffmpeg \
@@ -81,12 +71,11 @@ es.list.d/intelproducts.list && \
         git \
         graphviz \
         htop \
-        intel-mkl-2020.3-111 intel-ipp-2020.2-108 intel-tbb-2020.3-108 intel-da\
-al-2020.2-108 intel-mpi-2019.8-108 \
+        intel-mkl-2020.3-111 intel-ipp-2020.2-108 intel-tbb-2020.3-108 intel-daal-2020.2-108 intel-mpi-2019.8-108 \
         iputils-ping \
         libasound2-dev \
         libatlas-base-dev \
-	libavcodec-dev \
+	    libavcodec-dev \
         libavformat-dev \
         libavresample-dev \
         libboost-dev \
@@ -103,10 +92,10 @@ al-2020.2-108 intel-mpi-2019.8-108 \
         libgeos-dev \
         libgflags-dev \
         libgirepository1.0-dev \
-	libgoogle-glog-dev \
+	    libgoogle-glog-dev \
         libgphoto2-dev \
         libgstreamer-plugins-base1.0-dev \
-	libgstreamer1.0-dev \
+	    libgstreamer1.0-dev \
         libgtk-3-dev \
         libhdf5-dev \
         libhdf5-serial-dev \
@@ -134,8 +123,8 @@ al-2020.2-108 intel-mpi-2019.8-108 \
         libtbb-dev \
         libtheora-dev \
         libtiff-dev \
-                libtheora-dev \
-	libtiff-dev \
+        libtheora-dev \
+	    libtiff-dev \
         libtool \
         libv4l-dev \
         libvorbis-dev \
@@ -154,8 +143,8 @@ al-2020.2-108 intel-mpi-2019.8-108 \
         ncurses-term \
         net-tools \
         nodejs \
-#       nvtop \
-	openjdk-8-jdk \
+        nvtop \
+    	openjdk-8-jdk \
         openssh-client \
         openssh-server \
         pdsh \
@@ -172,10 +161,10 @@ al-2020.2-108 intel-mpi-2019.8-108 \
         yasm \
         zip \
         openssh-client \
-	build-essential \
+	    build-essential \
         gcc \
         git \
-	libavformat-dev \
+	    libavformat-dev \
         libcurl3-dev \
         libfreetype6-dev \
         libgeos-dev libgeos++-dev \
@@ -200,7 +189,6 @@ al-2020.2-108 intel-mpi-2019.8-108 \
     echo "/opt/intel/mkl/lib/intel64" >> /etc/ls.so.conf && \
     find /usr/local/cuda-${CUDA}/lib64/ -type f -name 'lib*_static.a' -not -name 'libcudart_static.a' -delete
 
-ENV MLNX_OFED_VERSION=5.4-3.1.0.0
 # Install OFED
 WORKDIR /tmp
 RUN wget -q -O - https://www.mellanox.com/downloads/ofed/MLNX_OFED-5.4-3.1.0.0\
@@ -223,22 +211,18 @@ RUN mkdir -p /opt/oracle && \
     apt-get clean && \
     rm -rf /var/lib/apt/lists/* && \
     apt-get update -y && \
-    apt-get install -y --no-install-recommends libaio1 default-libmysqlclient-d\
-ev && \
-    wget https://download.oracle.com/otn_software/linux/instantclient/instantcl\
-ient-basiclite-linuxx64.zip && \
+    apt-get install -y --no-install-recommends libaio1 default-libmysqlclient-dev && \
+    wget https://download.oracle.com/otn_software/linux/instantclient/instantclient-basiclite-linuxx64.zip && \
     unzip instantclient-basiclite-linuxx64.zip && \
     rm -f instantclient-basiclite-linuxx64.zip && \
     cd /opt/oracle/instantclient* && \
     rm -f *jdbc* *occi* *mysql* *README *jar uidrvci genezi adrci && \
-    echo /opt/oracle/instantclient* > /etc/ld.so.conf.d/oracle-instantclient.co\
-nf && \
+    echo /opt/oracle/instantclient* > /etc/ld.so.conf.d/oracle-instantclient.conf && \
     ldconfig
 
 # install git-lfs
 WORKDIR /tmp
-RUN curl -sLO https://github.com/git-lfs/git-lfs/releases/download/v3.0.2/git-l\
-fs-linux-amd64-v3.0.2.tar.gz && \
+RUN curl -sLO https://github.com/git-lfs/git-lfs/releases/download/v3.0.2/git-lfs-linux-amd64-v3.0.2.tar.gz && \
     tar -zxf git-lfs-linux-amd64-v3.0.2.tar.gz && \
     bash install.sh && \
     rm -rf /tmp/*
@@ -251,23 +235,18 @@ RUN git clone https://github.com/aristocratos/bashtop.git && \
 
 # Install VSCode
 WORKDIR /tmp
-RUN curl -fL https://github.com/cdr/code-server/releases/download/v3.12.0/code-\
-server-3.12.0-linux-amd64.tar.gz | tar -C /usr/local/lib -xz && \
-    mv /usr/local/lib/code-server-3.12.0-linux-amd64 /usr/local/lib/code-server\
--3.12.0 && \
-    ln -s /usr/local/lib/code-server-3.12.0/bin/code-server /usr/local/bin/code\
--server
+RUN curl -fL https://github.com/cdr/code-server/releases/download/v3.12.0/code-server-3.12.0-linux-amd64.tar.gz | tar -C /usr/local/lib -xz && \
+    mv /usr/local/lib/code-server-3.12.0-linux-amd64 /usr/local/lib/code-server-3.12.0 && \
+    ln -s /usr/local/lib/code-server-3.12.0/bin/code-server /usr/local/bin/code-server
 
 RUN rm /usr/bin/python3 && \
     ln -s /usr/bin/python3.6 /usr/bin/python3 && \
     update-alternatives --install /usr/bin/python python /usr/bin/python3 1 && \
-\
     update-alternatives --install /usr/bin/python3 python3 /usr/bin/python3.6 2
 
 WORKDIR /tmp
 RUN curl https://bootstrap.pypa.io/get-pip.py | python3 && \
-    python3 -m pip install --no-cache-dir -U setuptools pip Cython pybind11 pyt\
-hran
+    python3 -m pip install --no-cache-dir -U setuptools pip Cython pybind11 pythran
 
 # OpenBlas
 WORKDIR /tmp
@@ -283,26 +262,21 @@ py.git numpy && \
     git submodule update --init && \
     cp site.cfg.example site.cfg && \
     echo "[mkl]" >> site.cfg && \
-    echo "include_dirs = /opt/intel/compilers_and_libraries_2020.3.279/linux/mk\
-l/include/" >> site.cfg && \
-    echo "library_dirs = /opt/intel/compilers_and_libraries_2020.3.279/linux/mk\
-l/lib/intel64" >> site.cfg && \
+    echo "include_dirs = /opt/intel/compilers_and_libraries_2020.3.279/linux/mkl/include/" >> site.cfg && \
+    echo "library_dirs = /opt/intel/compilers_and_libraries_2020.3.279/linux/mkl/lib/intel64" >> site.cfg && \
     echo "mkl_libs = mkl_rt" >> site.cfg && \
     echo "lapack_libs =" >> site.cfg && \
     python3 setup.py bdist_wheel -d /tmp
 
 # Scipy
 RUN cd /tmp && \
-    git clone --branch=v${SCIPY_VERSION} --depth=1 https://github.com/scipy/sci\
-py.git scipy && \
+    git clone --branch=v${SCIPY_VERSION} --depth=1 https://github.com/scipy/scipy.git scipy && \
     cd scipy && \
     git checkout -b v${SCIPY_VERSION} && \
     cp site.cfg.example site.cfg && \
     echo "[mkl]" >> site.cfg && \
-    echo "include_dirs = /opt/intel/compilers_and_libraries_2020.3.279/linux/mk\
-l/include/" >> site.cfg && \
-    echo "library_dirs = /opt/intel/compilers_and_libraries_2020.3.279/linux/mk\
-l/lib/intel64" >> site.cfg && \
+    echo "include_dirs = /opt/intel/compilers_and_libraries_2020.3.279/linux/mkl/include/" >> site.cfg && \
+    echo "library_dirs = /opt/intel/compilers_and_libraries_2020.3.279/linux/mkl/lib/intel64" >> site.cfg && \
     echo "mkl_libs = mkl_rt" >> site.cfg && \
     echo "lapack_libs =" >> site.cfg && \
     python3 -m pip install -U --no-cache-dir /tmp/numpy-*.whl && \
@@ -314,8 +288,7 @@ RUN cd /tmp && \
 
 WORKDIR /tmp
 COPY ./requirements.py36.txt /tmp/requiremets.txt
-RUN python3 -m pip install --extra-index-url https://pypi.ngc.nvidia.com nvidia\
--pyindex && \
+RUN python3 -m pip install --extra-index-url https://pypi.ngc.nvidia.com nvidia-pyindex && \
     python3 -m pip install onnx-graphsurgeon && \
     python3 -m pip install /tmp/*.whl && \
     python3 -m pip install --no-cache-dir -r requirements.txt && \
@@ -325,8 +298,7 @@ RUN python3 -m pip install --extra-index-url https://pypi.ngc.nvidia.com nvidia\
 ENV OPENCV_VERSION="4.5.3"
 WORKDIR /tmp
 RUN wget https://github.com/opencv/opencv/archive/${OPENCV_VERSION}.zip && \
-    wget -O opencv-contrib.zip https://github.com/opencv/opencv_contrib/archive\
-/${OPENCV_VERSION}.zip && \
+    wget -O opencv-contrib.zip https://github.com/opencv/opencv_contrib/archive/${OPENCV_VERSION}.zip && \
     unzip ${OPENCV_VERSION}.zip && \
     unzip opencv-contrib.zip && \
     mkdir opencv-${OPENCV_VERSION}/cmake_binary && \
@@ -353,10 +325,8 @@ RUN wget https://github.com/opencv/opencv/archive/${OPENCV_VERSION}.zip && \
       -D CMAKE_BUILD_TYPE=RELEASE \
       -D CMAKE_INSTALL_PREFIX=$(python3 -c "import sys; print(sys.prefix)") \
       -D PYTHON_EXECUTABLE=$(which python3) \
-      -D PYTHON_INCLUDE_DIR=$(python3 -c "from distutils.sysconfig import get_p\
-ython_inc; print(get_python_inc())") \
-      -D PYTHON_PACKAGES_PATH=$(python3 -c "from distutils.sysconfig import get\
-_python_lib; print(get_python_lib())") \
+      -D PYTHON_INCLUDE_DIR=$(python3 -c "from distutils.sysconfig import get_python_inc; print(get_python_inc())") \
+      -D PYTHON_PACKAGES_PATH=$(python3 -c "from distutils.sysconfig import get_python_lib; print(get_python_lib())") \
       .. 2>&1 | tee cmake_messages.txt && \
     make -j$(nproc) && \
     make install && \
@@ -376,13 +346,10 @@ RUN git clone  --recursive https://github.com/dmlc/xgboost && \
 
 # LightGBM
 RUN cd /usr/local/src && mkdir lightgbm && cd lightgbm && \
-    git clone --recursive --branch stable --depth 1 https://github.com/microsof\
-t/LightGBM && \
+    git clone --recursive --branch stable --depth 1 https://github.com/microsoft/LightGBM && \
     cd LightGBM && mkdir build && cd build && \
-    cmake -DUSE_GPU=1 -DOpenCL_LIBRARY=/usr/local/cuda/lib64/libOpenCL.so -DOpe\
-nCL_INCLUDE_DIR=/usr/local/cuda/include/ .. && \
-    make OPENCL_HEADERS=/usr/local/cuda-11.3/targets/x86_64-linux/include LIBOP\
-ENCL=/usr/local/cuda-11.3/targets/x86_64-linux/lib -j$(nproc) && \
+    cmake -DUSE_GPU=1 -DOpenCL_LIBRARY=/usr/local/cuda/lib64/libOpenCL.so -DOpenCL_INCLUDE_DIR=/usr/local/cuda/include/ .. && \
+    make OPENCL_HEADERS=/usr/local/cuda-11.3/targets/x86_64-linux/include LIBOPENCL=/usr/local/cuda-11.3/targets/x86_64-linux/lib -j$(nproc) && \
     cd /usr/local/src/lightgbm/LightGBM/python-package && \
     python3 setup.py install --precompile
     # Install Open UCX
@@ -400,31 +367,25 @@ RUN wget https://github.com/openucx/ucx/archive/v1.10.0.tar.gz && \
 # Install Open MPI
 RUN mkdir /tmp/openmpi && \
     cd /tmp/openmpi && \
-    wget https://download.open-mpi.org/release/open-mpi/v4.1/openmpi-${OPENMPI}\
-.tar.gz && \
+    wget https://download.open-mpi.org/release/open-mpi/v4.1/openmpi-${OPENMPI}.tar.gz && \
     tar zxf openmpi-${OPENMPI}.tar.gz && \
     cd openmpi-${OPENMPI} && \
-    ./configure --enable-orterun-prefix-by-default --with-cuda --with-ucx=/usr/\
-local/ucx --enable-mca-no-build=btl-uct && \
+    ./configure --enable-orterun-prefix-by-default --with-cuda --with-ucx=/usr/local/ucx --enable-mca-no-build=btl-uct && \
     make -j $(nproc) all && \
     make install && \
     ldconfig && \
     rm -rf /tmp/openmpi*
 
 RUN python3 -m pip install --extra-index-url \
-       https://developer.download.nvidia.com/compute/redist nvidia-dali-cuda110\
- \
-       && \
+       https://developer.download.nvidia.com/compute/redist nvidia-dali-cuda110 && \
     python3 -m pip install --extra-index-url \
-       https://developer.download.nvidia.com/compute/redist nvidia-dali-tf-plug\
-in-cuda110
+       https://developer.download.nvidia.com/compute/redist nvidia-dali-tf-plugin-cuda110
 
 # Install OpenSSH for MPI to communicate between containers
 
 # Allow OpenSSH to talk to containers without asking for confirmation
 RUN mkdir -p /var/run/sshd && \
-    cat /etc/ssh/ssh_config | grep -v StrictHostKeyChecking > /etc/ssh/ssh_conf\
-ig.new && \
+    cat /etc/ssh/ssh_config | grep -v StrictHostKeyChecking > /etc/ssh/ssh_config.new && \
     echo "    StrictHostKeyChecking no" >> /etc/ssh/ssh_config.new && \
     mv /etc/ssh/ssh_config.new /etc/ssh/ssh_config
 
@@ -435,15 +396,12 @@ RUN mv /usr/local/bin/mpirun /usr/local/bin/mpirun.real && \
     chmod a+x /usr/local/bin/mpirun
 
 # Configure OpenMPI to run good defaults:
-RUN echo "btl_tcp_if_exclude = lo,docker0" >> /usr/local/etc/openmpi-mca-params\
-.conf
+RUN echo "btl_tcp_if_exclude = lo,docker0" >> /usr/local/etc/openmpi-mca-params.conf
 
 # Install Horovod, temporarily using CUDA stubs
 RUN ldconfig /usr/local/cuda/targets/x86_64-linux/lib/stubs && \
-    HOROVOD_GPU_ALLREDUCE=NCCL HOROVOD_GPU_BROADCAST=NCCL HOROVOD_NCCL_LINK=SHA\
-RED \
-    HOROVOD_WITH_TENSORFLOW=1 HOROVOD_WITHOUT_PYTORCH=1 HOROVOD_WITHOUT_MXNET=1\
-\
+    HOROVOD_GPU_ALLREDUCE=NCCL HOROVOD_GPU_BROADCAST=NCCL HOROVOD_NCCL_LINK=SHARED \
+    HOROVOD_WITH_TENSORFLOW=1 HOROVOD_WITHOUT_PYTORCH=1 HOROVOD_WITHOUT_MXNET=1 \
     HOROVOD_GPU=CUDA \
     python3 -m pip install --no-cache-dir horovod==0.22.1 && \
     ldconfig
@@ -456,8 +414,7 @@ RUN jupyter nbextensions_configurator enable && \
     jupyter contrib nbextension install && \
     jupyter nbextension enable --py --sys-prefix widgetsnbextension && \
     jupyter serverextension enable --py jupyterlab --sys-prefix && \
-    jupyter labextension install --no-build @jupyter-widgets/jupyterlab-manager\
- && \
+    jupyter labextension install --no-build @jupyter-widgets/jupyterlab-manager && \
     jupyter labextension install --no-build @krassowski/jupyterlab-lsp && \
     jupyter serverextension enable --py jupyter_lsp && \
     jupyter labextension install --no-build @jupyterlab/toc && \
@@ -468,20 +425,17 @@ RUN jupyter nbextensions_configurator enable && \
 
 RUN apt autoclean && \
     sed -i 's/source \/usr\/local\/nvm\/nvm.sh//' /etc/bash.bashrc && \
-    ln -sf /usr/share/terminfo/x/xterm-color /usr/share/terminfo/x/xterm-256col\
-or && \
+    ln -sf /usr/share/terminfo/x/xterm-color /usr/share/terminfo/x/xterm-256color && \
     rm -rf /var/lib/apt/lists/* && \
     rm -rf /root/.cache && \
     rm -rf /tmp/*
 # Install ipython kernelspec
-RUN /usr/bin/python3 -m ipykernel install --display-name "Python 3.6 (TensorFlo\
-w 2.5) on Backend.AI" && \
+RUN /usr/bin/python3 -m ipykernel install --display-name "Python 3.6 on Backend.AI" && \
     cat /usr/local/share/jupyter/kernels/python3/kernel.json
 
 # Backend.AI specifics
 COPY ./service-defs /etc/backend.ai/service-defs
-COPY runner-scripts/bootstrap.sh runner-scripts/setup_multinode.py /opt/contain\
-er/
+COPY runner-scripts/bootstrap.sh runner-scripts/setup_multinode.py /opt/container/
 LABEL ai.backend.kernelspec="1" \
       ai.backend.envs.corecount="OPENBLAS_NUM_THREADS,OMP_NUM_THREADS,NPROC" \
       ai.backend.features="batch query uid-match user-input" \
@@ -494,9 +448,6 @@ LABEL ai.backend.kernelspec="1" \
       ai.backend.base-distro="ubuntu16.04" \
       ai.backend.runtime-type="python" \
       ai.backend.runtime-path="/usr/bin/python" \
-      ai.backend.service-ports="ipython:pty:3000,jupyter:http:8091,jupyterlab:h\
-ttp:8090,vscode:http:8180,tensorboard:http:6006"
-
-
+      ai.backend.service-ports="ipython:pty:3000,jupyter:http:8091,jupyterlab:http:8090,vscode:http:8180,tensorboard:http:6006"
 
 WORKDIR /home/work