split up container into different parts for modularity

Dockerfile

@@ -1,98 +1,74 @@
-# bring in the micromamba image so we can copy files from it
-FROM mambaorg/micromamba:1.5.8 as micromamba
-
-# This is the image we are going add micromaba to:
-FROM cp2k/cp2k:2023.2_openmpi_generic_psmp
-
-USER root
-
-# if your image defaults to a non-root user, then you may want to make the
-# next 3 ARG commands match the values in your image. You can get the values
-# by running: docker run --rm -it my/image id -a
-ARG MAMBA_USER=mambauser
-ARG MAMBA_USER_ID=57439
-ARG MAMBA_USER_GID=57439
-ENV MAMBA_USER=$MAMBA_USER
-ENV MAMBA_ROOT_PREFIX="/opt/conda"
-ENV MAMBA_EXE="/bin/micromamba"
-
-COPY --from=micromamba "$MAMBA_EXE" "$MAMBA_EXE"
-COPY --from=micromamba /usr/local/bin/_activate_current_env.sh /usr/local/bin/_activate_current_env.sh
-COPY --from=micromamba /usr/local/bin/_dockerfile_shell.sh /usr/local/bin/_dockerfile_shell.sh
-COPY --from=micromamba /usr/local/bin/_entrypoint.sh /usr/local/bin/_entrypoint.sh
-COPY --from=micromamba /usr/local/bin/_dockerfile_initialize_user_accounts.sh /usr/local/bin/_dockerfile_initialize_user_accounts.sh
-COPY --from=micromamba /usr/local/bin/_dockerfile_setup_root_prefix.sh /usr/local/bin/_dockerfile_setup_root_prefix.sh
-
-RUN apt-get update
-RUN apt-get install ca-certificates git -y
-RUN update-ca-certificates
-
-RUN /usr/local/bin/_dockerfile_initialize_user_accounts.sh && \
-    /usr/local/bin/_dockerfile_setup_root_prefix.sh
-
-# modify entrypoint to also activate cp2k, but do not set up the included (lib)torch
-RUN head -n -1 /usr/local/bin/_entrypoint.sh > /usr/local/bin/entry.sh
-RUN sed '/torch/Id' /opt/cp2k/tools/toolchain/install/setup > /opt/cp2k/tools/toolchain/install/setup_notorch
-RUN echo "source /opt/cp2k/tools/toolchain/install/setup_notorch\n" >> /usr/local/bin/entry.sh
-RUN echo "export PATH=\"/opt/cp2k/exe/local:\${PATH}\"\n" >> /usr/local/bin/entry.sh
-
-RUN echo "exec \"\$@\"" >> /usr/local/bin/entry.sh
-RUN chmod +x /usr/local/bin/entry.sh
-
-USER $MAMBA_USER
-
-SHELL ["/usr/local/bin/_dockerfile_shell.sh"]
-
-ENTRYPOINT ["/usr/local/bin/entry.sh"]
-# Optional: if you want to customize the ENTRYPOINT and have a conda
-# environment activated, then do this:
-# ENTRYPOINT ["/usr/local/bin/_entrypoint.sh", "my_entrypoint_program"]
-
-# You can modify the CMD statement as needed....
-CMD ["/bin/bash"]
-
-RUN micromamba install -n base --yes -c conda-forge \
-    python=3.10 pip ndcctools py-plumed && \
-    micromamba clean -af --yes
-ARG MAMBA_DOCKERFILE_ACTIVATE=1  # (otherwise python will not be found)
-
-RUN pip install --no-cache-dir wandb plotly plumed 
-RUN pip install --no-cache-dir git+https://github.com/lab-cosmo/i-pi.git@feat/socket_prefix
+FROM ubuntu:22.04
+
+ENV DEBIAN_FRONTEND=noninteractive
+ENV TZ=Etc/UTC
+
+RUN apt-get update && apt-get install -y \
+    python3.10 \
+    python3.10-venv \
+    python3.10-dev \
+    build-essential \
+    cmake \
+    wget \
+    git \
+    zlib1g-dev \
+    libssl-dev \
+    libcurl4-openssl-dev \
+    libgsl-dev \
+    perl \
+    && apt-get clean \
+    && rm -rf /var/lib/apt/lists/*
+
+RUN python3.10 -m venv /opt/venv \
+    && /opt/venv/bin/pip install --upgrade pip
+
+# Install PLUMED
+ARG PLUMED_VERSION
+RUN wget https://github.com/plumed/plumed2/archive/refs/tags/v${PLUMED_VERSION}.tar.gz \
+    && tar -xzf v${PLUMED_VERSION}.tar.gz \
+    && cd plumed2-${PLUMED_VERSION} \
+    && ./configure --prefix=/usr/local/plumed \
+    && make -j$(nproc) \
+    && make install \
+    && cd .. \
+    && rm -rf plumed2-${PLUMED_VERSION} v${PLUMED_VERSION}.tar.gz
+
+# Ensure cctools can find the Python environment
+ENV PYTHONPATH="/opt/venv/lib/python3.10/site-packages:$PYTHONPATH"
+ENV PATH="/opt/venv/bin:$PATH"
+
+# Install cctools
+ARG CCTOOLS_VERSION
+RUN wget https://github.com/cooperative-computing-lab/cctools/archive/refs/tags/release/${CCTOOLS_VERSION}.tar.gz \
+    && tar -xzf ${CCTOOLS_VERSION}.tar.gz \
+    && cd cctools-release-${CCTOOLS_VERSION} \
+    && ./configure --prefix=/usr/local/cctools \
+    && make -j$(nproc) \
+    && make install \
+    && cd .. \
+    && rm -rf cctools-release-${CCTOOLS_VERSION} ${CCTOOLS_VERSION}.tar.gz
+
+# Set environment variables for PLUMED and cctools
+ENV PATH="/usr/local/plumed/bin:/usr/local/cctools/bin:$PATH"
+ENV LD_LIBRARY_PATH="/usr/local/plumed/lib:/usr/local/cctools/lib:$LD_LIBRARY_PATH"
+
+# Create entrypoint script
+RUN echo '#!/bin/bash\nsource /opt/venv/bin/activate\nexec "$@"' > /usr/local/bin/entrypoint.sh
+RUN chmod +x /usr/local/bin/entrypoint.sh
+
+ARG PSIFLOW_VERSION
+ARG PARSL_VERSION
 ARG GPU_LIBRARY
-RUN pip install --no-cache-dir torch==2.1 --index-url https://download.pytorch.org/whl/${GPU_LIBRARY}
-RUN pip install --no-cache-dir git+https://github.com/acesuit/[email protected]
-ARG GIT_COMMIT_SHA
-RUN pip install --no-cache-dir git+https://github.com/molmod/psiflow.git@ipi
-
-
-# install GPAW with MPI from cp2k
-#USER root
-#RUN git clone https://gitlab.com/libxc/libxc.git &&  \
-#cd libxc &&  \
-#apt install autogen autoconf libtool && \
-#autoreconf -i && \
-#./configure && \
-#make && \
-#make check && \
-#make install
-USER root
-RUN apt install libxc-dev -y
-USER $MAMBA_USER
-RUN source /opt/cp2k/tools/toolchain/install/setup && \
-export C_INCLUDE_PATH=$CPATH && \
-export LIBRARY_PATH=$LD_LIBRARY_PATH && \
-echo "C INCLUDE PATH" $C_INCLUDE_PATH && \
-pip install gpaw
-USER root
-RUN mkdir /opt/gpaw-data
-RUN yes | gpaw install-data /opt/gpaw-data || true
-ENV GPAW_SETUP_PATH=/opt/gpaw-data
-USER $MAMBA_USER
-
-#RUN echo "export GPAW_SETUP_PATH=\"/opt/gpaw-data\n" >> /usr/local/bin/entry.sh
-
-#ENV C_INCLUDE_PATH=$CPATH
-#RUN echo "export C_INCLUDE_PATH=\${CPATH}\"\n" >> /usr/local/bin/entry.sh
-#RUN /usr/local/bin/entry.sh
-#RUN echo "source /opt/cp2k/tools/toolchain/install/setup_notorch\n" >> /usr/local/bin/entry_gpaw.sh
-#RUN echo "source /opt/cp2k/tools/toolchain/install/setup_notorch\n" >> /usr/local/bin/entry_gpaw.sh
+RUN /bin/bash -c -o pipefail \
+    "source /opt/venv/bin/activate && \
+     pip install --no-cache-dir wandb plotly plumed && \
+     pip install --no-cache-dir git+https://github.com/lab-cosmo/i-pi.git@feat/socket_prefix && \
+     pip install --no-cache-dir git+https://github.com/molmod/psiflow.git@${PSIFLOW_VERSION} && \
+     pip install --no-cache-dir torch==2.1 --index-url https://download.pytorch.org/whl/${GPU_LIBRARY} && \
+     pip install --no-cache-dir git+https://github.com/acesuit/[email protected]"
+
+# Set entrypoint
+ENTRYPOINT ["/usr/local/bin/entrypoint.sh"]
+
+# Default command
+CMD ["bash"]

Dockerfile.cp2k

@@ -0,0 +1,110 @@
+#
+# This file was created by generate_docker_files.py
+#
+# Usage: docker build -f ./2023.2_openmpi_generic_psmp.Dockerfile -t cp2k/cp2k:2023.2_openmpi_generic_psmp .
+
+# Stage 1: build step
+FROM ubuntu:22.04 AS build
+
+
+# Install packages required for the CP2K toolchain build
+RUN apt-get update -qq && apt-get install -qq --no-install-recommends \
+    g++ gcc gfortran openssh-client python3 \
+    bzip2 ca-certificates git make patch pkg-config unzip wget zlib1g-dev
+
+# Download CP2K
+RUN git clone --recursive -b support/v2023.2 https://github.com/cp2k/cp2k.git /opt/cp2k
+
+# Build CP2K toolchain for target CPU generic
+WORKDIR /opt/cp2k/tools/toolchain
+RUN /bin/bash -c -o pipefail \
+    "./install_cp2k_toolchain.sh -j 8 \
+     --install-all \
+     --enable-cuda=no \
+     --target-cpu=generic \
+     --with-cusolvermp=no \
+     --with-gcc=system \
+     --with-openmpi=install \
+     --with-libtorch=no"
+
+# Build CP2K for target CPU generic
+WORKDIR /opt/cp2k
+RUN /bin/bash -c -o pipefail \
+    "cp ./tools/toolchain/install/arch/local.psmp ./arch/; \
+     source ./tools/toolchain/install/setup; \
+     make -j 8 ARCH=local VERSION=psmp"
+
+# Collect components for installation and remove symbolic links
+RUN /bin/bash -c -o pipefail \
+    "mkdir -p /toolchain/install /toolchain/scripts; \
+     for libdir in \$(ldd ./exe/local/cp2k.psmp | \
+                      grep /opt/cp2k/tools/toolchain/install | \
+                      awk '{print \$3}' | cut -d/ -f7 | \
+                      sort | uniq) setup; do \
+        cp -ar /opt/cp2k/tools/toolchain/install/\${libdir} /toolchain/install; \
+     done; \
+     cp /opt/cp2k/tools/toolchain/scripts/tool_kit.sh /toolchain/scripts; \
+     unlink ./exe/local/cp2k.popt; \
+     unlink ./exe/local/cp2k_shell.psmp"
+
+# Stage 2: install step
+FROM ubuntu:22.04 AS install
+
+# Install required packages
+RUN apt-get update -qq && apt-get install -qq --no-install-recommends \
+    g++ gcc gfortran openssh-client python3 && rm -rf /var/lib/apt/lists/*
+
+# Install CP2K binaries
+COPY --from=build /opt/cp2k/exe/local/ /opt/cp2k/exe/local/
+
+# Install CP2K regression tests
+COPY --from=build /opt/cp2k/tests/ /opt/cp2k/tests/
+COPY --from=build /opt/cp2k/tools/regtesting/ /opt/cp2k/tools/regtesting/
+COPY --from=build /opt/cp2k/src/grid/sample_tasks/ /opt/cp2k/src/grid/sample_tasks/
+
+# Install CP2K database files
+COPY --from=build /opt/cp2k/data/ /opt/cp2k/data/
+
+# Install shared libraries required by the CP2K binaries
+COPY --from=build /toolchain/ /opt/cp2k/tools/toolchain/
+
+# Create links to CP2K binaries
+RUN /bin/bash -c -o pipefail \
+    "for binary in cp2k dumpdcd graph xyz2dcd; do \
+        ln -sf /opt/cp2k/exe/local/\${binary}.psmp \
+               /usr/local/bin/\${binary}; \
+     done; \
+     ln -sf /opt/cp2k/exe/local/cp2k.psmp \
+            /usr/local/bin/cp2k_shell; \
+     ln -sf /opt/cp2k/exe/local/cp2k.psmp \
+            /usr/local/bin/cp2k.popt"
+
+# Create entrypoint script file
+RUN printf "#!/bin/bash\n\
+ulimit -c 0 -s unlimited\n\
+\
+export OMPI_ALLOW_RUN_AS_ROOT=1\n\
+export OMPI_ALLOW_RUN_AS_ROOT_CONFIRM=1\n\
+export OMPI_MCA_btl_vader_single_copy_mechanism=none\n\
+export OMP_STACKSIZE=16M\n\
+export PATH=/opt/cp2k/exe/local:\${PATH}\n\
+source /opt/cp2k/tools/toolchain/install/setup\n\
+\"\$@\"" \
+>/usr/local/bin/entrypoint.sh && chmod 755 /usr/local/bin/entrypoint.sh
+
+# Create shortcut for regression test
+RUN printf "/opt/cp2k/tools/regtesting/do_regtest.py --mpiexec \"mpiexec --bind-to none\" --maxtasks 8 --workbasedir /mnt \$* local psmp" \
+>/usr/local/bin/run_tests && chmod 755 /usr/local/bin/run_tests
+
+# Define entrypoint
+WORKDIR /mnt
+ENTRYPOINT ["/usr/local/bin/entrypoint.sh"]
+CMD ["cp2k", "--help"]
+
+# Label docker image
+LABEL author="CP2K Developers" \
+      cp2k_version="2023.2" \
+      dockerfile_generator_version="0.2"
+
+# EOF
+

Dockerfile.gpaw

@@ -0,0 +1,80 @@
+FROM ubuntu:22.04
+
+ENV DEBIAN_FRONTEND=noninteractive
+ENV TZ=Etc/UTC
+
+RUN apt-get update && apt-get install -y \
+    python3.10 \
+    python3.10-venv \
+    python3.10-dev \
+    build-essential \
+    autoconf \
+    automake \
+    libtool \
+    make \
+    cmake \
+    wget \
+    git \
+    libopenblas-dev \
+    libfftw3-dev \
+    libopenmpi-dev \
+    openmpi-bin \
+    libscalapack-mpi-dev \
+    libelpa-dev \
+    libomp-dev \
+    && apt-get clean \
+    && rm -rf /var/lib/apt/lists/*
+
+RUN python3.10 -m venv /opt/venv \
+    && /opt/venv/bin/pip install --upgrade pip
+
+RUN /bin/bash -c -o pipefail \
+    "source /opt/venv/bin/activate && \
+     pip install --no-cache-dir numpy && \
+     pip install --no-cache-dir git+https://gitlab.com/ase/ase"
+
+# install libxc
+RUN cd /opt && \
+    wget https://gitlab.com/libxc/libxc/-/archive/6.2.2/libxc-6.2.2.tar.bz2 -O libxc-6.2.2.tar.bz2 && \
+    tar -xvf libxc-6.2.2.tar.bz2 && \
+    cd libxc-6.2.2 && \
+    autoreconf -i && \
+    ./configure --prefix=/usr/local CFLAGS="-fPIC" && \
+    make -j$(nproc) && \
+    make install && \
+    ldconfig
+
+# install GPAW
+WORKDIR /opt/gpaw
+ENV GPAW_CONFIG=/opt/gpaw/siteconfig.py
+RUN echo "scalapack = True" >> siteconfig.py \
+    && echo "fftw = True" >> siteconfig.py \
+    && echo "elpa = True" >> siteconfig.py \
+    && echo "libraries = ['openblas', 'fftw3', 'scalapack-openmpi', 'elpa', 'omp5']" >> siteconfig.py \
+    && echo "library_dirs = ['/usr/lib', '/usr/local/lib']" >> siteconfig.py \
+    && echo "extra_link_args = ['/usr/local/lib/libxc.a']" >> siteconfig.py \
+    && echo "include_dirs = ['/usr/include', '/usr/local/include', '/usr/include/elpa']" >> siteconfig.py
+RUN /bin/bash -c -o pipefail "source /opt/venv/bin/activate && pip install --no-cache-dir gpaw dftd3"
+RUN mkdir /opt/gpaw-data
+RUN /bin/bash -c -o pipefail "source /opt/venv/bin/activate && yes | gpaw install-data /opt/gpaw-data" || true
+
+ARG PSIFLOW_VERSION
+RUN /bin/bash -c -o pipefail \
+    "source /opt/venv/bin/activate && pip install --no-cache-dir git+https://github.com/molmod/psiflow.git@${PSIFLOW_VERSION}"
+RUN ln -s /opt/venv/lib/python3.10/site-packages/psiflow/reference/gpaw_.py /opt/run_gpaw.py 
+
+# Create entrypoint script
+RUN echo '#!/bin/bash' >> /opt/entry.sh && \
+    echo 'source /opt/venv/bin/activate' >> /opt/entry.sh && \
+    echo 'export LD_LIBRARY_PATH="/usr/local/lib:${LD_LIBRARY_PATH}"' >> /opt/entry.sh && \
+    echo 'export GPAW_SETUP_PATH="/opt/gpaw-data/gpaw-setups-24.1.0"' >> /opt/entry.sh && \
+    echo '"$@"' >> /opt/entry.sh
+RUN chmod +x /opt/entry.sh
+
+# libxc needed at runtime
+ENV LD_LIBRARY_PATH="/usr/local/lib:${LD_LIBRARY_PATH}"
+ENTRYPOINT ["/opt/entry.sh"]
+
+
+# Default command
+CMD ["bash"]