Merge branch 'branch-24.12' into add-telemetry

.github/workflows/build.yaml

@@ -113,46 +113,6 @@ jobs:
       sha: ${{ inputs.sha }}
       date: ${{ inputs.date }}
       package-name: cugraph
-  wheel-build-cugraph-dgl:
-    needs: wheel-publish-cugraph
-    secrets: inherit
-    uses: rapidsai/shared-workflows/.github/workflows/[email protected]
-    with:
-      build_type: ${{ inputs.build_type || 'branch' }}
-      branch: ${{ inputs.branch }}
-      sha: ${{ inputs.sha }}
-      date: ${{ inputs.date }}
-      script: ci/build_wheel_cugraph-dgl.sh
-  wheel-publish-cugraph-dgl:
-    needs: wheel-build-cugraph-dgl
-    secrets: inherit
-    uses: rapidsai/shared-workflows/.github/workflows/[email protected]
-    with:
-      build_type: ${{ inputs.build_type || 'branch' }}
-      branch: ${{ inputs.branch }}
-      sha: ${{ inputs.sha }}
-      date: ${{ inputs.date }}
-      package-name: cugraph-dgl
-  wheel-build-cugraph-pyg:
-    needs: wheel-publish-cugraph
-    secrets: inherit
-    uses: rapidsai/shared-workflows/.github/workflows/[email protected]
-    with:
-      build_type: ${{ inputs.build_type || 'branch' }}
-      branch: ${{ inputs.branch }}
-      sha: ${{ inputs.sha }}
-      date: ${{ inputs.date }}
-      script: ci/build_wheel_cugraph-pyg.sh
-  wheel-publish-cugraph-pyg:
-    needs: wheel-build-cugraph-pyg
-    secrets: inherit
-    uses: rapidsai/shared-workflows/.github/workflows/[email protected]
-    with:
-      build_type: ${{ inputs.build_type || 'branch' }}
-      branch: ${{ inputs.branch }}
-      sha: ${{ inputs.sha }}
-      date: ${{ inputs.date }}
-      package-name: cugraph-pyg
   wheel-build-cugraph-equivariant:
     secrets: inherit
     uses: rapidsai/shared-workflows/.github/workflows/[email protected]

.github/workflows/pr.yaml

@@ -33,10 +33,6 @@ jobs:
       - wheel-tests-pylibcugraph
       - wheel-build-cugraph
       - wheel-tests-cugraph
-      - wheel-build-cugraph-dgl
-      - wheel-tests-cugraph-dgl
-      - wheel-build-cugraph-pyg
-      - wheel-tests-cugraph-pyg
       - wheel-build-cugraph-equivariant
       - wheel-tests-cugraph-equivariant
       - telemetry-setup
@@ -177,38 +173,6 @@ jobs:
     with:
       build_type: pull-request
       script: ci/test_wheel_cugraph.sh
-  wheel-build-cugraph-dgl:
-    needs: wheel-tests-cugraph
-    secrets: inherit
-    uses: rapidsai/shared-workflows/.github/workflows/[email protected]
-    with:
-      build_type: pull-request
-      script: ci/build_wheel_cugraph-dgl.sh
-  wheel-tests-cugraph-dgl:
-    needs: [wheel-build-cugraph-dgl, changed-files]
-    secrets: inherit
-    uses: rapidsai/shared-workflows/.github/workflows/[email protected]
-    if: fromJSON(needs.changed-files.outputs.changed_file_groups).test_python
-    with:
-      build_type: pull-request
-      script: ci/test_wheel_cugraph-dgl.sh
-      matrix_filter: map(select(.ARCH == "amd64"))
-  wheel-build-cugraph-pyg:
-    needs: wheel-tests-cugraph
-    secrets: inherit
-    uses: rapidsai/shared-workflows/.github/workflows/[email protected]
-    with:
-      build_type: pull-request
-      script: ci/build_wheel_cugraph-pyg.sh
-  wheel-tests-cugraph-pyg:
-    needs: [wheel-build-cugraph-pyg, changed-files]
-    secrets: inherit
-    uses: rapidsai/shared-workflows/.github/workflows/[email protected]
-    if: fromJSON(needs.changed-files.outputs.changed_file_groups).test_python
-    with:
-      build_type: pull-request
-      script: ci/test_wheel_cugraph-pyg.sh
-      matrix_filter: map(select(.ARCH == "amd64"))
   wheel-build-cugraph-equivariant:
     secrets: inherit
     uses: rapidsai/shared-workflows/.github/workflows/[email protected]

.github/workflows/test.yaml

@@ -58,26 +58,6 @@ jobs:
       date: ${{ inputs.date }}
       sha: ${{ inputs.sha }}
       script: ci/test_wheel_cugraph.sh
-  wheel-tests-cugraph-dgl:
-    secrets: inherit
-    uses: rapidsai/shared-workflows/.github/workflows/[email protected]
-    with:
-      build_type: nightly
-      branch: ${{ inputs.branch }}
-      date: ${{ inputs.date }}
-      sha: ${{ inputs.sha }}
-      script: ci/test_wheel_cugraph-dgl.sh
-      matrix_filter: map(select(.ARCH == "amd64"))
-  wheel-tests-cugraph-pyg:
-    secrets: inherit
-    uses: rapidsai/shared-workflows/.github/workflows/[email protected]
-    with:
-      build_type: nightly
-      branch: ${{ inputs.branch }}
-      date: ${{ inputs.date }}
-      sha: ${{ inputs.sha }}
-      script: ci/test_wheel_cugraph-pyg.sh
-      matrix_filter: map(select(.ARCH == "amd64"))
   wheel-tests-cugraph-equivariant:
     secrets: inherit
     uses: rapidsai/shared-workflows/.github/workflows/[email protected]

ci/build_python.sh

@@ -50,29 +50,6 @@ rapids-conda-retry mambabuild \
   --channel "${RAPIDS_CONDA_BLD_OUTPUT_DIR}" \
   conda/recipes/cugraph-service
 
-RAPIDS_CUDA_MAJOR="${RAPIDS_CUDA_VERSION%%.*}"
-
-if [[ ${RAPIDS_CUDA_MAJOR} == "11" ]]; then
-  # Only CUDA 11 is supported right now due to PyTorch requirement.
-  rapids-conda-retry mambabuild \
-    --no-test \
-    --channel "${CPP_CHANNEL}" \
-    --channel "${RAPIDS_CONDA_BLD_OUTPUT_DIR}" \
-    --channel pytorch \
-    --channel pytorch-nightly \
-    conda/recipes/cugraph-pyg
-
-  # Only CUDA 11 is supported right now due to PyTorch requirement.
-  rapids-conda-retry mambabuild \
-    --no-test \
-    --channel "${CPP_CHANNEL}" \
-    --channel "${RAPIDS_CONDA_BLD_OUTPUT_DIR}" \
-    --channel dglteam/label/th23_cu118 \
-    --channel pytorch \
-    --channel pytorch-nightly \
-    conda/recipes/cugraph-dgl
-fi
-
 rapids-conda-retry mambabuild \
   --no-test \
   --channel "${RAPIDS_CONDA_BLD_OUTPUT_DIR}" \

ci/build_wheel.sh

@@ -30,9 +30,7 @@ python -m pip wheel \
 sccache --show-adv-stats
 
 # pure-python packages should be marked as pure, and not have auditwheel run on them.
-if [[ ${package_name} == "cugraph-dgl" ]] || \
-   [[ ${package_name} == "cugraph-pyg" ]] || \
-   [[ ${package_name} == "cugraph-equivariant" ]]; then
+if [[ ${package_name} == "cugraph-equivariant" ]]; then
     RAPIDS_PY_WHEEL_NAME="${package_name}_${RAPIDS_PY_CUDA_SUFFIX}" RAPIDS_PY_WHEEL_PURE="1" rapids-upload-wheels-to-s3 python dist
 else
     case "${RAPIDS_CUDA_VERSION}" in

ci/build_wheel_cugraph-equivariant.sh

@@ -3,4 +3,7 @@
 
 set -euo pipefail
 
-./ci/build_wheel.sh cugraph-equivariant python/cugraph-equivariant
+package_dir="python/cugraph-equivariant"
+
+./ci/build_wheel.sh cugraph-equivariant ${package_dir}
+./ci/validate_wheel.sh ${package_dir} dist

ci/build_wheel_cugraph.sh

@@ -3,6 +3,8 @@
 
 set -euo pipefail
 
+package_dir="python/cugraph"
+
 RAPIDS_PY_CUDA_SUFFIX="$(rapids-wheel-ctk-name-gen ${RAPIDS_CUDA_VERSION})"
 
 # Download the pylibcugraph wheel built in the previous step and make it
@@ -31,4 +33,5 @@ esac
 export SKBUILD_CMAKE_ARGS="-DDETECT_CONDA_ENV=OFF;-DFIND_CUGRAPH_CPP=OFF;-DCPM_cugraph-ops_SOURCE=${GITHUB_WORKSPACE}/cugraph-ops/${EXTRA_CMAKE_ARGS}"
 export SKBUILD_BUILD_TOOL_ARGS="-j${PARALLEL_LEVEL};-l${PARALLEL_LEVEL}"
 
-./ci/build_wheel.sh cugraph python/cugraph
+./ci/build_wheel.sh cugraph ${package_dir}
+./ci/validate_wheel.sh ${package_dir} final_dist

ci/build_wheel_pylibcugraph.sh

@@ -3,6 +3,8 @@
 
 set -euo pipefail
 
+package_dir="python/pylibcugraph"
+
 PARALLEL_LEVEL=$(python -c \
   "from math import ceil; from multiprocessing import cpu_count; print(ceil(cpu_count()/4))")
 
@@ -18,4 +20,5 @@ esac
 export SKBUILD_CMAKE_ARGS="-DDETECT_CONDA_ENV=OFF;-DFIND_CUGRAPH_CPP=OFF;-DCPM_cugraph-ops_SOURCE=${GITHUB_WORKSPACE}/cugraph-ops/${EXTRA_CMAKE_ARGS}"
 export SKBUILD_BUILD_TOOL_ARGS="-j${PARALLEL_LEVEL};-l${PARALLEL_LEVEL}"
 
-./ci/build_wheel.sh pylibcugraph python/pylibcugraph
+./ci/build_wheel.sh pylibcugraph ${package_dir}
+./ci/validate_wheel.sh ${package_dir} final_dist

ci/test_python.sh

@@ -100,109 +100,12 @@ rapids-logger "pytest cugraph-service (single GPU)"
   --cov-report=xml:"${RAPIDS_COVERAGE_DIR}/cugraph-service-coverage.xml" \
   --cov-report=term
 
-if [[ "${RAPIDS_CUDA_VERSION}" == "11.8.0" ]]; then
-  if [[ "${RUNNER_ARCH}" != "ARM64" ]]; then
-    # we are only testing in a single cuda version
-    # because of pytorch and rapids compatibilty problems
-    rapids-mamba-retry env create --yes -f env.yaml -n test_cugraph_dgl
-
-    # activate test_cugraph_dgl environment for dgl
-    set +u
-    conda activate test_cugraph_dgl
-    set -u
-    rapids-mamba-retry install \
-      --channel "${CPP_CHANNEL}" \
-      --channel "${PYTHON_CHANNEL}" \
-      --channel conda-forge \
-      --channel dglteam/label/th23_cu118 \
-      --channel nvidia \
-      "libcugraph=${RAPIDS_VERSION_MAJOR_MINOR}.*" \
-      "pylibcugraph=${RAPIDS_VERSION_MAJOR_MINOR}.*" \
-      "pylibcugraphops=${RAPIDS_VERSION_MAJOR_MINOR}.*" \
-      "cugraph=${RAPIDS_VERSION_MAJOR_MINOR}.*" \
-      "cugraph-dgl=${RAPIDS_VERSION_MAJOR_MINOR}.*" \
-      'pytorch>=2.3,<2.4' \
-      'cuda-version=11.8'
-
-    rapids-print-env
-
-    rapids-logger "pytest cugraph_dgl (single GPU)"
-    ./ci/run_cugraph_dgl_pytests.sh \
-      --junitxml="${RAPIDS_TESTS_DIR}/junit-cugraph-dgl.xml" \
-      --cov-config=../../.coveragerc \
-      --cov=cugraph_dgl \
-      --cov-report=xml:"${RAPIDS_COVERAGE_DIR}/cugraph-dgl-coverage.xml" \
-      --cov-report=term
-
-    # Reactivate the test environment back
-    set +u
-    conda deactivate
-    conda activate test
-    set -u
-  else
-    rapids-logger "skipping cugraph_dgl pytest on ARM64"
-  fi
-else
-  rapids-logger "skipping cugraph_dgl pytest on CUDA!=11.8"
-fi
-
-if [[ "${RAPIDS_CUDA_VERSION}" == "11.8.0" ]]; then
-  if [[ "${RUNNER_ARCH}" != "ARM64" ]]; then
-    rapids-mamba-retry env create --yes -f env.yaml -n test_cugraph_pyg
-
-    # Temporarily allow unbound variables for conda activation.
-    set +u
-    conda activate test_cugraph_pyg
-    set -u
-
-    rapids-print-env
-
-    # TODO re-enable logic once CUDA 12 is testable
-    #if [[ "${RAPIDS_CUDA_VERSION}" == "11.8.0" ]]; then
-    CONDA_CUDA_VERSION="11.8"
-    PYG_URL="https://data.pyg.org/whl/torch-2.3.0+cu118.html"
-    #else
-    #  CONDA_CUDA_VERSION="12.1"
-    #  PYG_URL="https://data.pyg.org/whl/torch-2.3.0+cu121.html"
-    #fi
-
-    # Will automatically install built dependencies of cuGraph-PyG
-    rapids-mamba-retry install \
-      --channel "${CPP_CHANNEL}" \
-      --channel "${PYTHON_CHANNEL}" \
-      "cugraph-pyg=${RAPIDS_VERSION_MAJOR_MINOR}.*" \
-      "pytorch>=2.3,<2.4" \
-      "ogb"
-
-    rapids-print-env
-
-    rapids-logger "pytest cugraph_pyg (single GPU)"
-    # rmat is not tested because of multi-GPU testing
-    ./ci/run_cugraph_pyg_pytests.sh \
-      --junitxml="${RAPIDS_TESTS_DIR}/junit-cugraph-pyg.xml" \
-      --cov-config=../../.coveragerc \
-      --cov=cugraph_pyg \
-      --cov-report=xml:"${RAPIDS_COVERAGE_DIR}/cugraph-pyg-coverage.xml" \
-      --cov-report=term
-
-    # Reactivate the test environment back
-    set +u
-    conda deactivate
-    conda activate test
-    set -u
-  else
-    rapids-logger "skipping cugraph_pyg pytest on ARM64"
-  fi
-else
-  rapids-logger "skipping cugraph_pyg pytest on CUDA!=11.8"
-fi
-
 # test cugraph-equivariant
 if [[ "${RAPIDS_CUDA_VERSION}" == "11.8.0" ]]; then
   if [[ "${RUNNER_ARCH}" != "ARM64" ]]; then
-    # Reuse cugraph-dgl's test env for cugraph-equivariant
+    rapids-mamba-retry env create --yes -f env.yaml -n test_cugraph_equivariant
     set +u
-    conda activate test_cugraph_dgl
+    conda activate test_cugraph_equivariant
     set -u
     rapids-mamba-retry install \
       --channel "${CPP_CHANNEL}" \