heddlemd/Containerfile at main · MolSSI/heddlemd · GitHub

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
ARG RIPRAP_AGENT_IMAGE
FROM ${RIPRAP_AGENT_IMAGE}
#FROM docker.io/nvidia/cuda:11.8.0-devel-ubuntu22.04

ENV DEBIAN_FRONTEND=noninteractive

# The 11.8 base toolkit predates Blackwell (RTX 50-series, compute capability
# 12.0 / sm_120) by two GPU generations: its nvcc/NVRTC only know up to
# compute_90, so HeddleMD's runtime-JITed pair/SPME kernels fail with
# "invalid value for --gpu-architecture (-arch)" on an sm_120 device. Install a
# Blackwell-capable toolkit (13.2, matching the driver's CUDA 13.2 support).
# update-alternatives repoints /usr/local/cuda -> /usr/local/cuda-13.2, so nvcc
# on PATH and the cudarc-loaded libnvrtc/libcufft are all 13.2.
# (Cleaner long-term alternative: switch the FROM base image to an
# nvidia/cuda:13.2.x-devel-ubuntu22.04 tag instead of layering the toolkit.)
RUN apt-get update && \
    apt-get install -y --no-install-recommends cuda-toolkit-13-2 && \
    apt-get clean && \
    rm -rf /var/lib/apt/lists/*
ENV CUDA_PATH=/usr/local/cuda
ENV CUDA_HOME=/usr/local/cuda
ENV PATH="/usr/local/cuda/bin:${PATH}"
ENV LD_LIBRARY_PATH="/usr/local/cuda/lib64:${LD_LIBRARY_PATH}"

RUN apt-get clean && \
    apt-get update && \
    apt-get install -y software-properties-common && \
    apt-get update && \
    apt-get install -y --no-install-recommends \
                       clang \
                       clangd \
                       cmake \
                       curl \
                       gcc \
                       gdb \
                       git \
                       make \
                       npm \
                       python3-venv \
                       python3-pip \
                       vim && \
    apt-get clean && \
    rm -rf /var/lib/apt/lists/*

# Add python
RUN apt-get clean && \
    apt-get update && \
    apt-get install -y \
                       python3 \
                       python3-pip \
                       python3-dev \
                       && \
    apt-get clean && \
    rm -rf /var/lib/apt/lists/*

# Symlink python3 to python
RUN ln -s /usr/bin/python3 /usr/bin/python

# Install PyCUDA and other Python packages
RUN python -m pip install --upgrade pip setuptools wheel && \
    python -m pip install \
                          pycuda \
                          numpy \
                          scipy \
                          matplotlib \
                          nvidia-cutlass-dsl==4.2.0 \
                          pandas \
                          jupyter


ENV PATH="$PATH:/root/.local/bin"

# Install Rust
ENV RUST_VERSION=1.93.0
RUN curl https://sh.rustup.rs -sSf | sh -s -- -y --default-toolchain ${RUST_VERSION}
ENV PATH="/root/.cargo/bin:${PATH}"

# Install mdBook for building user-facing documentation
RUN cargo install mdbook --locked


COPY .podman/interface.sh /interface.sh
RUN mkdir -p /.podman
COPY .podman/interface.sh /.podman/interface.sh

# Copy the entrypoint files into the Docker image
COPY .podman/entrypoint.sh /.term/entrypoint.sh
RUN chmod +x /.term/entrypoint.sh

# Set the default command
CMD ["/.term/entrypoint.sh"]