rocprofiler-compute Docker Images in GHCR (#1195)

* Initial cleanup of compute workflows and skeleton of ghcr workflow

* Add containers-ci.yml, update opensuse and rhel dockerfiles

* rename id in rocprofiler-compute-ghcr.yml

* Add new line to end of containers-ci.yml

* Update action versions for rocprofiler-compute-ghcr.yml

* Switch back to SHA for action versions

* Add conda set solver classic fix to compute CI dockerfiles

* Update conda install for compute Dockerfiles

* Change opensuse version to 15.6 in containers-ci.yml

* Add fix for ubuntu noble to compute Dockerfile.ubuntu.ci

* Add default distro and version to Dockerfile.ubuntu.ci

* Updated regex for tarball version

* Remove Python3.8 from compute CI Dockerfiles

* Change RHEL 9.4 to 9, add retry for compute workflow

* Revert name change for compute rhel workflow

* update path naming

* Remove binutils-gold from Dockerfile.opensuse.ci

* Remove conda python installs from Dockerfile.ci files in compute

* Change CMake version to 3.21 in compute Dockerfile.ci files

* Update checkout actions from v4 to v5

.github/workflows/rocprofiler-compute-ghcr.yml

@@ -0,0 +1,136 @@
+name: Publish GHCR Packages for rocprofiler-compute CI Images
+
+on:
+  workflow_dispatch:
+  schedule:
+    - cron: 0 5 * * *
+  push:
+    branches:
+      - develop
+    paths:
+      - '.github/workflows/rocprofiler-compute-ghcr.yml'
+      - 'projects/rocprofiler-compute/docker/**'
+  pull_request:
+    paths:
+      - '.github/workflows/rocprofiler-compute-ghcr.yml'
+      - 'projects/rocprofiler-compute/docker/**'
+
+jobs:
+  prepare_matrix_ci:
+    if: github.repository == 'ROCm/rocm-systems'
+    runs-on: ubuntu-latest
+    outputs:
+      matrix_data: ${{ steps.generate_matrix_ci.outputs.matrix_data }}
+
+    steps:
+      - uses: actions/checkout@v5
+        with:
+          sparse-checkout: projects/rocprofiler-compute/docker
+
+      - name: Output data for containers matrix
+        working-directory: projects/rocprofiler-compute/docker
+        id: generate_matrix_ci
+        run: |
+          MATRIX_CONTENT=$(cat containers-ci.yml | yq '.matrix' -I=0 -o=json)
+          echo "matrix_data=$MATRIX_CONTENT" >> $GITHUB_OUTPUT
+
+  rocprofiler-compute-ci-gfx:
+    needs: prepare_matrix_ci
+    runs-on: ubuntu-latest
+    permissions:
+      contents: read
+      packages: write
+
+    strategy:
+      fail-fast: false
+      matrix:
+        system: ${{ fromJSON(needs.prepare_matrix_ci.outputs.matrix_data) }}
+        gpu: [ 'gfx94X', 'gfx950' ]
+
+    steps:
+      - name: Checkout Repository
+        uses: actions/checkout@v5
+        with:
+          sparse-checkout: projects/rocprofiler-compute
+          submodules: recursive
+
+      - name: Login to GitHub Container Registry
+        uses: docker/login-action@5e57cd118135c172c3672efd75eb46360885c0ef # v3.6.0
+        with:
+          registry: ghcr.io
+          username: ${{ github.actor }}
+          password: ${{ secrets.GITHUB_TOKEN }}
+
+      - name: Set up Docker variables
+        id: setup_vars_gfx
+        run: |
+          if [ ${{ matrix.system.distro }} = "opensuse" ]; then
+            DISTRO_IMAGE="opensuse/leap"
+          elif [ ${{ matrix.system.distro }} = "rhel" ]; then
+            DISTRO_IMAGE="rockylinux/rockylinux"
+          else
+            DISTRO_IMAGE=${{ matrix.system.distro }}
+          fi
+          echo "distro_image=${DISTRO_IMAGE}" >> $GITHUB_OUTPUT
+
+          if [ ${{ matrix.system.distro }} = "debian" ]; then
+            DOCKER_FILE=Dockerfile.ubuntu.ci
+          else
+            DOCKER_FILE=Dockerfile.${{ matrix.system.distro }}.ci
+          fi
+          echo "docker_file=${DOCKER_FILE}" >> $GITHUB_OUTPUT
+      
+      - name: Get the latest build of The Rock tarball
+        id: therock
+        run: |
+          sudo apt-get install -y python3-pip
+          python3 -m pip install -U pip
+          python3 -m pip install -U awscli
+          export PATH=~/.local/bin:$PATH
+          KEY=$(aws s3api list-objects-v2 \
+                --bucket therock-nightly-tarball \
+                --no-sign-request \
+                --output json \
+                --query "sort_by(Contents[?contains(Key, 'linux-${{ matrix.gpu }}')], &LastModified)[-1].Key")
+          KEY=${KEY//\"/}
+          test -n "$KEY" || { echo "No ${{ matrix.gpu }} tarball found"; exit 1; }
+          echo "tarball=${KEY}" >> $GITHUB_OUTPUT
+      
+      - name: Extract metadata (tags, labels) for Docker
+        id: meta
+        uses: docker/metadata-action@c1e51972afc2121e065aed6d45c65596fe445f3f # v5.8.0
+        with:
+          images: ghcr.io/ROCm/rocprofiler-${{ matrix.system.distro }}
+
+      - name: Build CI GFX Container (Does not Push on PR)
+        id: docker_build
+        uses: docker/build-push-action@263435318d21b8e681c14492fe198d362a7d2c83 # v6.18.0
+        with:
+          file: projects/rocprofiler-compute/docker/${{ steps.setup_vars_gfx.outputs.docker_file }}
+          platforms: linux/amd64
+          push: ${{ github.event_name != 'pull_request' }}
+          build-args: |
+            DISTRO=${{ steps.setup_vars_gfx.outputs.distro_image }}
+            VERSION=${{ matrix.system.version }}
+            GPU_TYPE=${{ matrix.gpu }}
+            GPU_TARBALL=${{ steps.therock.outputs.tarball }}
+          tags: |
+            ghcr.io/rocm/rocprofiler-${{ matrix.system.distro }}:${{ matrix.system.version }}-compute-ci-${{ matrix.gpu }}
+          labels: ${{ steps.meta.outputs.labels }}
+
+      # Retry a copy of docker_build if Docker build failed due to intermittent failure
+      - name: Build CI GFX Container Retry (Does not Push on PR)
+        if: steps.docker_build.outcome != 'success'
+        uses: docker/build-push-action@263435318d21b8e681c14492fe198d362a7d2c83 # v6.18.0
+        with:
+          file: projects/rocprofiler-compute/docker/${{ steps.setup_vars_gfx.outputs.docker_file }}
+          platforms: linux/amd64
+          push: ${{ github.event_name != 'pull_request' }}
+          build-args: |
+            DISTRO=${{ steps.setup_vars_gfx.outputs.distro_image }}
+            VERSION=${{ matrix.system.version }}
+            GPU_TYPE=${{ matrix.gpu }}
+            GPU_TARBALL=${{ steps.therock.outputs.tarball }}
+          tags: |
+            ghcr.io/rocm/rocprofiler-${{ matrix.system.distro }}:${{ matrix.system.version }}-compute-ci-${{ matrix.gpu }}
+          labels: ${{ steps.meta.outputs.labels }}

.github/workflows/rocprofiler-compute-rhel-8.yml

@@ -1,8 +1,5 @@
-# This is a basic workflow to help you get started with Actions
+name: rocprofiler-compute RHEL
 
-name: rocprofiler-compute RHEL 8/9
-
-# Controls when the workflow will run
 on:
   push:
     branches:
@@ -32,8 +29,6 @@ on:
       - '!**/.wordlist.txt'
       - '!projects/rocprofiler-compute/docs/**'
       - '!projects/rocprofiler-compute/docker/**'
-
-  # Allows you to run this workflow manually from the Actions tab
   workflow_dispatch:
 
 concurrency:
@@ -43,9 +38,8 @@ concurrency:
 env:
   ROCM_VERSION: "7.2.0"
 
-# A workflow run is made up of one or more jobs that can run sequentially or in parallel
 jobs:
-  build:
+  build-rhel:
     # The type of runner that the job will run on
     # Using rocprofiler-systems images and later installing rocm, the official rocm images for al8 red hat variant are too large- this is a workaround.
     runs-on: ubuntu-latest
@@ -56,6 +50,7 @@ jobs:
       matrix:
         os-release: [ '8.10', '9.4' ]
         build-type: ['Release']
+
     # Steps represent a sequence of tasks that will be executed as part of the job
     # Note- RHEL8 references python3.9 as python39, whereas RHEL9 default version is already 3.9, so we only need to call it as python3
     steps:
@@ -75,6 +70,7 @@ jobs:
           yum -y install cmake3
           yum -y install which
           yum -y install glibc-langpack-en
+
       - name: Install ROCm Packages
         timeout-minutes: 30
         uses: nick-fields/retry@v3
@@ -92,10 +88,12 @@ jobs:
               yum install -y https://repo.radeon.com/amdgpu-install/${ROCM_MAJOR}.${ROCM_MINOR}/rhel/9.4/amdgpu-install-${ROCM_MAJOR}.${ROCM_MINOR}.${ROCM_VERSN}-1.el9.noarch.rpm
             fi
             yum install -y rocm-dev
+            
       - name: Checkout
-        uses: actions/checkout@v4
+        uses: actions/checkout@v5
         with:
           sparse-checkout: projects/rocprofiler-compute
+
       - name: Install Python prereqs
         working-directory: projects/rocprofiler-compute
         run: |
@@ -109,11 +107,13 @@ jobs:
           cd build
           cmake -DCMAKE_INSTALL_PREFIX=/opt/rocprofiler-compute -DPYTEST_NUMPROCS=4 ..
           make install
+
       - name: CTest- Analyze Commands
         working-directory: projects/rocprofiler-compute
         run: |
           cd build
           ctest --verbose -R test_analyze_commands
+
       - name: CTest- Analyze Workloads
         working-directory: projects/rocprofiler-compute
         run: |

.github/workflows/rocprofiler-compute-ubuntu-jammy.yml

@@ -1,5 +1,3 @@
-# This is a basic workflow to help you get started with Actions
-
 name: rocprofiler-compute Ubuntu 22.04
 
 on:
@@ -31,8 +29,6 @@ on:
       - '!**/.wordlist.txt'
       - '!projects/rocprofiler-compute/docs/**'
       - '!projects/rocprofiler-compute/docker/**'
-
-  # Allows you to run this workflow manually from the Actions tab
   workflow_dispatch:
 
 concurrency:
@@ -40,7 +36,7 @@ concurrency:
   cancel-in-progress: true
 
 jobs:
-  build:
+  build-ubuntu-jammy:
     # The type of runner that the job will run on
     # https://hub.docker.com/r/rocm/dev-ubuntu-22.04/tags
     runs-on: ubuntu-latest
@@ -54,15 +50,18 @@ jobs:
           apt-get install -y git
           apt-get install -y python3-pip
           apt-get install -y cmake
+
       - name: Checkout
-        uses: actions/checkout@v4
+        uses: actions/checkout@v5
         with:
           sparse-checkout: projects/rocprofiler-compute
+
       - name: Install Python prereqs
         working-directory: projects/rocprofiler-compute
         run: |
           python3 -m pip install -r requirements.txt
           python3 -m pip install -r requirements-test.txt
+
       - name: Configure and install
         working-directory: projects/rocprofiler-compute
         run: |
@@ -70,11 +69,13 @@ jobs:
           cd build
           cmake -DCMAKE_INSTALL_PREFIX=/opt/rocprofiler-compute -DPYTEST_NUMPROCS=4 ..
           make install
+
       - name: CTest- Analyze Commands
         working-directory: projects/rocprofiler-compute
         run: |
           cd build
           ctest --verbose -R test_analyze_commands
+
       - name: CTest- Analyze Workloads
         working-directory: projects/rocprofiler-compute
         run: |

projects/rocprofiler-compute/docker/Dockerfile.opensuse.ci

@@ -2,19 +2,17 @@ ARG DISTRO=opensuse/leap
 ARG VERSION=15.3
 FROM ${DISTRO}:${VERSION}
 
-ENV HOME /root
+ENV HOME=/root
-ENV SHELL /bin/bash
+ENV SHELL=/bin/bash
-ENV BASH_ENV /etc/bash.bashrc
+ENV BASH_ENV=/etc/bash.bashrc
-ENV DEBIAN_FRONTEND noninteractive
+ENV DEBIAN_FRONTEND=noninteractive
 
 WORKDIR /tmp
 SHELL [ "/bin/bash", "-c" ]
 
-ENV PATH /usr/local/bin:${PATH}
+ENV PATH=/usr/local/bin:${PATH}
 
 ARG EXTRA_PACKAGES=""
-# ARG ELFUTILS_DOWNLOAD_VERSION="0.186"
-# ARG NJOBS="12"
 
 RUN set +e; \
     zypper --non-interactive -i --gpg-auto-import-keys refresh; \
@@ -26,23 +24,23 @@ RUN set +e; \
 RUN zypper --non-interactive update -y && \
     zypper --non-interactive dist-upgrade -y && \
     zypper --non-interactive install -y -t pattern devel_basis && \
-    zypper --non-interactive install -y python3-pip gcc-c++ git dpkg-devel rpm-build curl wget binutils-gold && \
+    zypper --non-interactive install -y python3-pip gcc-c++ git dpkg-devel rpm-build curl wget && \
-    python3 -m pip install 'cmake==3.28.4' && \
+    python3 -m pip install 'cmake==3.21' && \
     zypper --non-interactive clean --all
 
-ARG PYTHON_VERSIONS="6 7 8 9 10 11 12"
+# The Rock Tarball
-
+ARG GPU_TYPE=""
-RUN wget https://repo.continuum.io/miniconda/Miniconda3-latest-Linux-x86_64.sh -O miniconda.sh && \
+ARG GPU_TARBALL=""
-    bash miniconda.sh -b -p /opt/conda && \
+RUN if [ -n "$GPU_TYPE" ] && [ -n "$GPU_TARBALL" ]; then \
-    export PATH="/opt/conda/bin:${PATH}" && \
+        VERSION=$(echo "$GPU_TARBALL" | sed -nE 's/.*([0-9]+\.[0-9]+\.[0-9]+).*/\1/p'); \
-    conda config --set always_yes yes --set changeps1 no && \
+        if [ -z "$VERSION" ]; then \
-    conda update -c defaults -n base conda && \
+            echo "Error: Could not extract version from GPU_TARBALL ('$GPU_TARBALL')." >&2; \
-    for i in ${PYTHON_VERSIONS}; do conda create -n py3.${i} -c defaults -c conda-forge python=3.${i} pip numpy; done && \
+            exit 1; \
-    for i in ${PYTHON_VERSIONS}; do /opt/conda/envs/py3.${i}/bin/python -m pip install numpy pandas dataclasses; done && \
+        fi; \
-    conda clean -a -y && \
+        python3 -m pip install -U awscli; \
-    cd /tmp && \
+        aws s3 cp "s3://therock-nightly-tarball/${GPU_TARBALL}" rocm-${VERSION}-${GPU_TYPE}.tar.gz --no-sign-request; \
-    shopt -s dotglob extglob && \
+        mv rocm-${VERSION}-${GPU_TYPE}.tar.gz /opt/rocm-${VERSION}-${GPU_TYPE}.tar.gz; \
-    rm -rf *
+    fi
 
 WORKDIR /home
 SHELL [ "/bin/bash", "--login", "-c" ]

projects/rocprofiler-compute/docker/Dockerfile.rhel.ci

@@ -1,42 +1,39 @@
-
 ARG DISTRO=rockylinux
 ARG VERSION=8
 FROM ${DISTRO}:${VERSION}
 
-ENV HOME /root
+ENV HOME=/root
-ENV SHELL /bin/bash
+ENV SHELL=/bin/bash
-ENV BASH_ENV /etc/bash.bashrc
+ENV BASH_ENV=/etc/bash.bashrc
-ENV DEBIAN_FRONTEND noninteractive
+ENV DEBIAN_FRONTEND=noninteractive
 
 WORKDIR /tmp
 SHELL [ "/bin/bash", "-c" ]
 
-ENV PATH /usr/local/bin:${PATH}
+ENV PATH=/usr/local/bin:${PATH}
 
 ARG EXTRA_PACKAGES=""
-# ARG ELFUTILS_DOWNLOAD_VERSION="0.186"
-# ARG NJOBS="12"
 
 RUN yum groupinstall -y "Development Tools" && \
     yum install -y epel-release && \
     yum install -y --allowerasing curl dpkg-devel python3-pip wget zlib-devel which git && \
     yum clean all && \
     python3 -m pip install --upgrade pip && \
-    python3 -m pip install 'cmake==3.28.4'
+    python3 -m pip install 'cmake==3.21'
 
-ARG PYTHON_VERSIONS="6 7 8 9 10 11 12"
+# The Rock Tarball
-
+ARG GPU_TYPE=""
-RUN wget https://repo.continuum.io/miniconda/Miniconda3-latest-Linux-x86_64.sh -O miniconda.sh && \
+ARG GPU_TARBALL=""
-    bash miniconda.sh -b -p /opt/conda && \
+RUN if [ -n "$GPU_TYPE" ] && [ -n "$GPU_TARBALL" ]; then \
-    export PATH="/opt/conda/bin:${PATH}" && \
+        VERSION=$(echo "$GPU_TARBALL" | sed -nE 's/.*([0-9]+\.[0-9]+\.[0-9]+).*/\1/p'); \
-    conda config --set always_yes yes --set changeps1 no && \
+        if [ -z "$VERSION" ]; then \
-    conda update -c defaults -n base conda && \
+            echo "Error: Could not extract version from GPU_TARBALL ('$GPU_TARBALL')." >&2; \
-    for i in ${PYTHON_VERSIONS}; do conda create -n py3.${i} -c defaults -c conda-forge python=3.${i} pip numpy; done && \
+            exit 1; \
-    for i in ${PYTHON_VERSIONS}; do /opt/conda/envs/py3.${i}/bin/python -m pip install numpy pandas dataclasses; done && \
+        fi; \
-    conda clean -a -y && \
+        python3 -m pip install -U awscli; \
-    cd /tmp && \
+        aws s3 cp "s3://therock-nightly-tarball/${GPU_TARBALL}" rocm-${VERSION}-${GPU_TYPE}.tar.gz --no-sign-request; \
-    shopt -s dotglob extglob && \
+        mv rocm-${VERSION}-${GPU_TYPE}.tar.gz /opt/rocm-${VERSION}-${GPU_TYPE}.tar.gz; \
-    rm -rf *
+    fi
 
 WORKDIR /home
 SHELL [ "/bin/bash", "--login", "-c" ]

projects/rocprofiler-compute/docker/Dockerfile.ubuntu.ci

@@ -1,48 +1,52 @@
-
+ARG DISTRO=ubuntu
-ARG DISTRO
+ARG VERSION=22.04
-ARG VERSION
 FROM ${DISTRO}:${VERSION}
 
-ENV HOME /root
+ENV HOME=/root
-ENV LANG C.UTF-8
+ENV LANG=C.UTF-8
-ENV SHELL /bin/bash
+ENV SHELL=/bin/bash
-ENV BASH_ENV /etc/bash.bashrc
+ENV BASH_ENV=/etc/bash.bashrc
-ENV DEBIAN_FRONTEND noninteractive
+ENV DEBIAN_FRONTEND=noninteractive
 
 WORKDIR /tmp
 SHELL [ "/bin/bash", "-c" ]
 
 ARG EXTRA_PACKAGES=""
-# ARG ELFUTILS_DOWNLOAD_VERSION="0.186"
-# ARG BOOST_DOWNLOAD_VERSION="1.79.0"
-# ARG NJOBS="12"
-ARG PYTHON_VERSIONS="6 7 8 9 10 11 12"
-
-ENV PATH /usr/local/bin:${PATH}
-ENV LIBRARY_PATH /usr/local/lib:/usr/local/lib64:${LIBRARY_PATH}
-ENV LD_LIBRARY_PATH /usr/local/lib:/usr/local/lib64:${LD_LIBRARY_PATH}
-ENV CMAKE_PREFIX_PATH /usr/local:${CMAKE_PREFIX_PATH}
 
+ENV PATH="/usr/local/bin:${PATH}"
+ENV LIBRARY_PATH="/usr/local/lib:/usr/local/lib64"
+ENV LD_LIBRARY_PATH="/usr/local/lib:/usr/local/lib64"
+ENV CMAKE_PREFIX_PATH="/usr/local"
 
 RUN apt-get update && \
     apt-get dist-upgrade -y && \
     apt-get install -y autoconf autotools-dev bash-completion build-essential bzip2 cmake curl environment-modules git-core gnupg2 gzip libtool locales lsb-release m4 python3-pip unzip wget zip zlib1g-dev && \
-    python3 -m pip install 'cmake==3.28.4' && \
     apt-get autoclean && \
-    locale -a && \
+    locale -a
-    wget https://repo.continuum.io/miniconda/Miniconda3-latest-Linux-x86_64.sh -O miniconda.sh && \
-    bash miniconda.sh -b -p /opt/conda && \
-    export PATH="/opt/conda/bin:${PATH}" && \
-    conda config --set always_yes yes --set changeps1 no && \
-    conda update -c defaults -n base conda && \
-    for i in ${PYTHON_VERSIONS}; do conda create -n py3.${i} -c defaults -c conda-forge python=3.${i} pip numpy; done && \
-    for i in ${PYTHON_VERSIONS}; do /opt/conda/envs/py3.${i}/bin/python -m pip install numpy pandas dataclasses; done && \
-    conda clean -a -y && \
-    cd /tmp && \
-    shopt -s dotglob extglob && \
-    rm -rf *
 
+RUN OS_VERSION=$(grep '^VERSION_ID=' /etc/os-release | cut -d'=' -f2 | tr -d '"') && \
+    OS_ID=$(grep '^ID=' /etc/os-release | cut -d'=' -f2 | tr -d '"') && \
+    if [ "${OS_ID}" == "ubuntu" ] && [ "${OS_VERSION}" == "22.04" ]; then \
+        python3 -m pip install 'cmake==3.21'; \
+    else \
+        python3 -m pip install --break-system-packages 'cmake==3.21'; \
+    fi
 
-ENV LC_ALL C.UTF-8
+# The Rock Tarball
+ARG GPU_TYPE=""
+ARG GPU_TARBALL=""
+RUN if [ -n "$GPU_TYPE" ] && [ -n "$GPU_TARBALL" ]; then \
+        VERSION=$(echo "$GPU_TARBALL" | sed -nE 's/.*([0-9]+\.[0-9]+\.[0-9]+).*/\1/p'); \
+        if [ -z "$VERSION" ]; then \
+            echo "Error: Could not extract version from GPU_TARBALL ('$GPU_TARBALL')." >&2; \
+            exit 1; \
+        fi; \
+        pip install --upgrade pip; \
+        pip install awscli --break-system-packages; \
+        aws s3 cp "s3://therock-nightly-tarball/${GPU_TARBALL}" rocm-${VERSION}-${GPU_TYPE}.tar.gz --no-sign-request; \
+        mv rocm-${VERSION}-${GPU_TYPE}.tar.gz /opt/rocm-${VERSION}-${GPU_TYPE}.tar.gz; \
+    fi
+
+ENV LC_ALL=C.UTF-8
 WORKDIR /home
 SHELL [ "/bin/bash", "--login", "-c" ]

projects/rocprofiler-compute/docker/containers-ci.yml

@@ -0,0 +1,17 @@
+# Supported OS combinations for continuous integration
+
+matrix:
+  # ubuntu
+  - distro: "ubuntu"
+    version: "22.04"
+  - distro: "ubuntu"
+    version: "24.04"
+  # opensuse
+  - distro: "opensuse"
+    version: "15.6"
+  # RHEL8
+  - distro: "rhel"
+    version: "8.10"
+  # RHEL9
+  - distro: "rhel"
+    version: "9"