adding wheel build for libcudf

.github/workflows/build.yaml

@@ -67,7 +67,27 @@ jobs:
       node_type: "gpu-v100-latest-1"
       run_script: "ci/build_docs.sh"
       sha: ${{ inputs.sha }}
+  wheel-build-libcudf:
+    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_libcudf.sh
+  wheel-publish-libcudf:
+    needs: wheel-build-libcudf
+    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: libcudf
   wheel-build-pylibcudf:
+    needs: [wheel-publish-libcudf]
     secrets: inherit
     uses: rapidsai/shared-workflows/.github/workflows/[email protected]
     with:

.github/workflows/pr.yaml

@@ -23,6 +23,7 @@ jobs:
       - static-configure
       - conda-notebook-tests
       - docs-build
+      - wheel-build-libcudf
       - wheel-build-pylibcudf
       - wheel-build-cudf
       - wheel-tests-cudf
@@ -121,10 +122,18 @@ jobs:
       arch: "amd64"
       container_image: "rapidsai/ci-conda:latest"
       run_script: "ci/build_docs.sh"
-  wheel-build-pylibcudf:
+  wheel-build-libcudf:
     needs: checks
     secrets: inherit
     uses: rapidsai/shared-workflows/.github/workflows/[email protected]
+    with:
+      matrix_filter: group_by([.ARCH, (.CUDA_VER|split(".")|map(tonumber)|.[0])]) | map(max_by(.PY_VER|split(".")|map(tonumber)))
+      build_type: pull-request
+      script: "ci/build_wheel_libcudf.sh"
+  wheel-build-pylibcudf:
+    needs: [checks, wheel-build-libcudf]
+    secrets: inherit
+    uses: rapidsai/shared-workflows/.github/workflows/[email protected]
     with:
       build_type: pull-request
       script: "ci/build_wheel_pylibcudf.sh"

ci/build_wheel_cudf.sh

@@ -5,16 +5,28 @@ set -euo pipefail
 
 package_dir="python/cudf"
 
-export SKBUILD_CMAKE_ARGS="-DUSE_LIBARROW_FROM_PYARROW=ON"
-
-# Download the pylibcudf built in the previous step
 RAPIDS_PY_CUDA_SUFFIX="$(rapids-wheel-ctk-name-gen ${RAPIDS_CUDA_VERSION})"
-RAPIDS_PY_WHEEL_NAME="pylibcudf_${RAPIDS_PY_CUDA_SUFFIX}" rapids-download-wheels-from-s3 /tmp/pylibcudf_dist
 
-echo "pylibcudf-${RAPIDS_PY_CUDA_SUFFIX} @ file://$(echo /tmp/pylibcudf_dist/pylibcudf_*.whl)" > /tmp/constraints.txt
+# Downloads libcudf and pylibcudf wheels from this current build,
+# then ensures 'cudf' wheel builds always use the 'libcudf' and 'pylibcudf' just built in the same CI run.
+#
+# Using env variable PIP_CONSTRAINT is necessary to ensure the constraints
+# are used when creating the isolated build environment.
+RAPIDS_PY_WHEEL_NAME="libcudf_${RAPIDS_PY_CUDA_SUFFIX}" rapids-download-wheels-from-s3 cpp /tmp/libcudf_dist
+RAPIDS_PY_WHEEL_NAME="pylibcudf_${RAPIDS_PY_CUDA_SUFFIX}" rapids-download-wheels-from-s3 python /tmp/pylibcudf_dist
+echo "libcudf-${RAPIDS_PY_CUDA_SUFFIX} @ file://$(echo /tmp/libcudf_dist/libcudf_*.whl)" > /tmp/constraints.txt
+echo "pylibcudf-${RAPIDS_PY_CUDA_SUFFIX} @ file://$(echo /tmp/pylibcudf_dist/pylibcudf_*.whl)" >> /tmp/constraints.txt
 export PIP_CONSTRAINT="/tmp/constraints.txt"
+
 ./ci/build_wheel.sh ${package_dir}
 
-python -m auditwheel repair -w ${package_dir}/final_dist ${package_dir}/dist/*
+python -m auditwheel repair \
+    --exclude libcudf.so \
+    --exclude libarrow.so.1601 \
+    --exclude libnvcomp.so \
+    --exclude libnvcomp_bitcomp.so \
+    --exclude libnvcomp_gdeflate.so \
+    -w ${package_dir}/final_dist \
+    ${package_dir}/dist/*
 
-RAPIDS_PY_WHEEL_NAME="cudf_${RAPIDS_PY_CUDA_SUFFIX}" rapids-upload-wheels-to-s3 ${package_dir}/final_dist
+RAPIDS_PY_WHEEL_NAME="cudf_${RAPIDS_PY_CUDA_SUFFIX}" rapids-upload-wheels-to-s3 python ${package_dir}/final_dist

ci/build_wheel_libcudf.sh

@@ -0,0 +1,15 @@
+#!/bin/bash
+# Copyright (c) 2023-2024, NVIDIA CORPORATION.
+
+set -euo pipefail
+
+package_dir="python/libcudf"
+
+./ci/build_wheel.sh ${package_dir}
+
+RAPIDS_PY_CUDA_SUFFIX="$(rapids-wheel-ctk-name-gen ${RAPIDS_CUDA_VERSION})"
+
+mkdir -p ${package_dir}/final_dist
+python -m auditwheel repair --exclude libarrow.so.1601 -w ${package_dir}/final_dist ${package_dir}/dist/*
+
+RAPIDS_PY_WHEEL_NAME="libcudf_${RAPIDS_PY_CUDA_SUFFIX}" rapids-upload-wheels-to-s3 cpp ${package_dir}/final_dist

ci/build_wheel_pylibcudf.sh

@@ -5,12 +5,26 @@ set -euo pipefail
 
 package_dir="python/pylibcudf"
 
-export SKBUILD_CMAKE_ARGS="-DUSE_LIBARROW_FROM_PYARROW=ON"
+RAPIDS_PY_CUDA_SUFFIX="$(rapids-wheel-ctk-name-gen ${RAPIDS_CUDA_VERSION})"
 
-./ci/build_wheel.sh ${package_dir}
+# Downloads libcudf wheel from this current build,
+# then ensures 'pylibcudf' wheel builds always use the 'libcudf' just built in the same CI run.
+#
+# Using env variable PIP_CONSTRAINT is necessary to ensure the constraints
+# are used when creating the isolated build environment.
+RAPIDS_PY_WHEEL_NAME="libcudf_${RAPIDS_PY_CUDA_SUFFIX}" rapids-download-wheels-from-s3 cpp /tmp/libcudf_dist
+echo "libcudf-${RAPIDS_PY_CUDA_SUFFIX} @ file://$(echo /tmp/libcudf_dist/libcudf_*.whl)" > /tmp/constraints.txt
+export PIP_CONSTRAINT="/tmp/constraints.txt"
 
-python -m auditwheel repair -w ${package_dir}/final_dist ${package_dir}/dist/*
+./ci/build_wheel.sh ${package_dir}
 
+python -m auditwheel repair \
+    --exclude libcudf.so \
+    --exclude libarrow.so.1601 \
+    --exclude libnvcomp.so \
+    --exclude libnvcomp_bitcomp.so \
+    --exclude libnvcomp_gdeflate.so \
+    -w ${package_dir}/final_dist \
+    ${package_dir}/dist/*
 
-RAPIDS_PY_CUDA_SUFFIX="$(rapids-wheel-ctk-name-gen ${RAPIDS_CUDA_VERSION})"
 RAPIDS_PY_WHEEL_NAME="pylibcudf_${RAPIDS_PY_CUDA_SUFFIX}" rapids-upload-wheels-to-s3 ${package_dir}/final_dist

ci/cudf_pandas_scripts/pandas-tests/run.sh

@@ -12,13 +12,15 @@ rapids-logger "PR number: ${RAPIDS_REF_NAME:-"unknown"}"
 
 RAPIDS_PY_CUDA_SUFFIX="$(rapids-wheel-ctk-name-gen ${RAPIDS_CUDA_VERSION})"
 
-# Download the cudf and pylibcudf built in the previous step
-RAPIDS_PY_WHEEL_NAME="cudf_${RAPIDS_PY_CUDA_SUFFIX}" rapids-download-wheels-from-s3 ./dist
-RAPIDS_PY_WHEEL_NAME="pylibcudf_${RAPIDS_PY_CUDA_SUFFIX}" rapids-download-wheels-from-s3 ./dist
+# Download the cudf, libcudf, and pylibcudf built in the previous step
+RAPIDS_PY_WHEEL_NAME="cudf_${RAPIDS_PY_CUDA_SUFFIX}" rapids-download-wheels-from-s3 python ./dist
+RAPIDS_PY_WHEEL_NAME="libcudf_${RAPIDS_PY_CUDA_SUFFIX}" rapids-download-wheels-from-s3 cpp ./dist
+RAPIDS_PY_WHEEL_NAME="pylibcudf_${RAPIDS_PY_CUDA_SUFFIX}" rapids-download-wheels-from-s3 python ./dist
 
 # echo to expand wildcard before adding `[extra]` requires for pip
 python -m pip install \
   "$(echo ./dist/cudf_${RAPIDS_PY_CUDA_SUFFIX}*.whl)[test,pandas-tests]" \
+  "$(echo ./dist/libcudf_${RAPIDS_PY_CUDA_SUFFIX}*.whl)" \
   "$(echo ./dist/pylibcudf_${RAPIDS_PY_CUDA_SUFFIX}*.whl)"
 
 RESULTS_DIR=${RAPIDS_TESTS_DIR:-"$(mktemp -d)"}

ci/cudf_pandas_scripts/run_tests.sh

@@ -37,13 +37,15 @@ if [ "$no_cudf" = true ]; then
 else
     RAPIDS_PY_CUDA_SUFFIX="$(rapids-wheel-ctk-name-gen ${RAPIDS_CUDA_VERSION})"
 
-    # Download the cudf and pylibcudf built in the previous step
-    RAPIDS_PY_WHEEL_NAME="cudf_${RAPIDS_PY_CUDA_SUFFIX}" rapids-download-wheels-from-s3 ./dist
-    RAPIDS_PY_WHEEL_NAME="pylibcudf_${RAPIDS_PY_CUDA_SUFFIX}" rapids-download-wheels-from-s3 ./dist
+    # Download the cudf, libcudf, and pylibcudf built in the previous step
+    RAPIDS_PY_WHEEL_NAME="cudf_${RAPIDS_PY_CUDA_SUFFIX}" rapids-download-wheels-from-s3 python ./dist
+    RAPIDS_PY_WHEEL_NAME="libcudf_${RAPIDS_PY_CUDA_SUFFIX}" rapids-download-wheels-from-s3 cpp ./dist
+    RAPIDS_PY_WHEEL_NAME="pylibcudf_${RAPIDS_PY_CUDA_SUFFIX}" rapids-download-wheels-from-s3 python ./dist
 
     # echo to expand wildcard before adding `[extra]` requires for pip
     python -m pip install \
         "$(echo ./dist/cudf_${RAPIDS_PY_CUDA_SUFFIX}*.whl)[test,cudf-pandas-tests]" \
+        "$(echo ./dist/libcudf_${RAPIDS_PY_CUDA_SUFFIX}*.whl)" \
         "$(echo ./dist/pylibcudf_${RAPIDS_PY_CUDA_SUFFIX}*.whl)"
 fi
 

ci/release/update-version.sh

@@ -49,6 +49,7 @@ DEPENDENCIES=(
   dask-cuda
   dask-cudf
   kvikio
+  libcudf
   libkvikio
   librmm
   pylibcudf

ci/test_wheel_cudf.sh

@@ -5,13 +5,15 @@ set -eou pipefail
 
 RAPIDS_PY_CUDA_SUFFIX="$(rapids-wheel-ctk-name-gen ${RAPIDS_CUDA_VERSION})"
 
-# Download the cudf and pylibcudf built in the previous step
-RAPIDS_PY_WHEEL_NAME="cudf_${RAPIDS_PY_CUDA_SUFFIX}" rapids-download-wheels-from-s3 ./dist
-RAPIDS_PY_WHEEL_NAME="pylibcudf_${RAPIDS_PY_CUDA_SUFFIX}" rapids-download-wheels-from-s3 ./dist
+# Download the cudf, libcudf, and pylibcudf built in the previous step
+RAPIDS_PY_WHEEL_NAME="cudf_${RAPIDS_PY_CUDA_SUFFIX}" rapids-download-wheels-from-s3 python ./dist
+RAPIDS_PY_WHEEL_NAME="libcudf_${RAPIDS_PY_CUDA_SUFFIX}" rapids-download-wheels-from-s3 cpp ./dist
+RAPIDS_PY_WHEEL_NAME="pylibcudf_${RAPIDS_PY_CUDA_SUFFIX}" rapids-download-wheels-from-s3 python ./dist
 
 # echo to expand wildcard before adding `[extra]` requires for pip
 python -m pip install \
   "$(echo ./dist/cudf_${RAPIDS_PY_CUDA_SUFFIX}*.whl)[test]" \
+  "$(echo ./dist/libcudf_${RAPIDS_PY_CUDA_SUFFIX}*.whl)" \
   "$(echo ./dist/pylibcudf_${RAPIDS_PY_CUDA_SUFFIX}*.whl)[test]"
 
 RESULTS_DIR=${RAPIDS_TESTS_DIR:-"$(mktemp -d)"}

ci/test_wheel_cudf_polars.sh

@@ -18,16 +18,18 @@ else
 fi
 
 RAPIDS_PY_CUDA_SUFFIX="$(rapids-wheel-ctk-name-gen ${RAPIDS_CUDA_VERSION})"
-RAPIDS_PY_WHEEL_NAME="cudf_polars_${RAPIDS_PY_CUDA_SUFFIX}" RAPIDS_PY_WHEEL_PURE="1" rapids-download-wheels-from-s3 ./dist
+RAPIDS_PY_WHEEL_NAME="cudf_polars_${RAPIDS_PY_CUDA_SUFFIX}" RAPIDS_PY_WHEEL_PURE="1" rapids-download-wheels-from-s3 python ./dist
 
-# Download pylibcudf built in the previous step
-RAPIDS_PY_WHEEL_NAME="pylibcudf_${RAPIDS_PY_CUDA_SUFFIX}" rapids-download-wheels-from-s3 ./dist
+# Download libcudf and pylibcudf built in the previous step
+RAPIDS_PY_WHEEL_NAME="libcudf_${RAPIDS_PY_CUDA_SUFFIX}" rapids-download-wheels-from-s3 cpp ./dist
+RAPIDS_PY_WHEEL_NAME="pylibcudf_${RAPIDS_PY_CUDA_SUFFIX}" rapids-download-wheels-from-s3 python ./dist
 
 rapids-logger "Installing cudf_polars and its dependencies"
 
 # echo to expand wildcard before adding `[extra]` requires for pip
 python -m pip install \
     "$(echo ./dist/cudf_polars_${RAPIDS_PY_CUDA_SUFFIX}*.whl)[test]" \
+    "$(echo ./dist/libcudf_${RAPIDS_PY_CUDA_SUFFIX}*.whl)" \
     "$(echo ./dist/pylibcudf_${RAPIDS_PY_CUDA_SUFFIX}*.whl)"
 
 rapids-logger "Run cudf_polars tests"

ci/test_wheel_dask_cudf.sh

@@ -4,16 +4,18 @@
 set -eou pipefail
 
 RAPIDS_PY_CUDA_SUFFIX="$(rapids-wheel-ctk-name-gen ${RAPIDS_CUDA_VERSION})"
-RAPIDS_PY_WHEEL_NAME="dask_cudf_${RAPIDS_PY_CUDA_SUFFIX}" RAPIDS_PY_WHEEL_PURE="1" rapids-download-wheels-from-s3 ./dist
+RAPIDS_PY_WHEEL_NAME="dask_cudf_${RAPIDS_PY_CUDA_SUFFIX}" RAPIDS_PY_WHEEL_PURE="1" rapids-download-wheels-from-s3 python ./dist
 
-# Download the cudf and pylibcudf built in the previous step
-RAPIDS_PY_WHEEL_NAME="cudf_${RAPIDS_PY_CUDA_SUFFIX}" rapids-download-wheels-from-s3 ./dist
-RAPIDS_PY_WHEEL_NAME="pylibcudf_${RAPIDS_PY_CUDA_SUFFIX}" rapids-download-wheels-from-s3 ./dist
+# Download the cudf, libcudf, and pylibcudf built in the previous step
+RAPIDS_PY_WHEEL_NAME="cudf_${RAPIDS_PY_CUDA_SUFFIX}" rapids-download-wheels-from-s3 python ./dist
+RAPIDS_PY_WHEEL_NAME="libcudf_${RAPIDS_PY_CUDA_SUFFIX}" rapids-download-wheels-from-s3 cpp ./dist
+RAPIDS_PY_WHEEL_NAME="pylibcudf_${RAPIDS_PY_CUDA_SUFFIX}" rapids-download-wheels-from-s3 python ./dist
 
 # echo to expand wildcard before adding `[extra]` requires for pip
 python -m pip install \
   "$(echo ./dist/cudf_${RAPIDS_PY_CUDA_SUFFIX}*.whl)" \
   "$(echo ./dist/dask_cudf_${RAPIDS_PY_CUDA_SUFFIX}*.whl)[test]" \
+  "$(echo ./dist/libcudf_${RAPIDS_PY_CUDA_SUFFIX}*.whl)" \
   "$(echo ./dist/pylibcudf_${RAPIDS_PY_CUDA_SUFFIX}*.whl)"
 
 RESULTS_DIR=${RAPIDS_TESTS_DIR:-"$(mktemp -d)"}