Merge branch 'branch-23.12' into fix-precision-decimal128

rapidsai · Oct 9, 2023 · 68d8ff4 · 68d8ff4
2 parents c1d023e + e345620
commit 68d8ff4
Show file tree

Hide file tree

Showing 84 changed files with 1,664 additions and 1,126 deletions.
diff --git a/.github/workflows/build.yaml b/.github/workflows/build.yaml
@@ -28,7 +28,7 @@ concurrency:
 jobs:
   cpp-build:
     secrets: inherit
-    uses: rapidsai/shared-action-workflows/.github/workflows/conda-cpp-build.yaml@branch-23.12
+    uses: rapidsai/shared-action-workflows/.github/workflows/conda-cpp-build.yaml@cuda-120-arm
     with:
       build_type: ${{ inputs.build_type || 'branch' }}
       branch: ${{ inputs.branch }}
@@ -37,7 +37,7 @@ jobs:
   python-build:
     needs: [cpp-build]
     secrets: inherit
-    uses: rapidsai/shared-action-workflows/.github/workflows/conda-python-build.yaml@branch-23.12
+    uses: rapidsai/shared-action-workflows/.github/workflows/conda-python-build.yaml@cuda-120-arm
     with:
       build_type: ${{ inputs.build_type || 'branch' }}
       branch: ${{ inputs.branch }}
@@ -46,7 +46,7 @@ jobs:
   upload-conda:
     needs: [cpp-build, python-build]
     secrets: inherit
-    uses: rapidsai/shared-action-workflows/.github/workflows/conda-upload-packages.yaml@branch-23.12
+    uses: rapidsai/shared-action-workflows/.github/workflows/conda-upload-packages.yaml@cuda-120-arm
     with:
       build_type: ${{ inputs.build_type || 'branch' }}
       branch: ${{ inputs.branch }}
@@ -57,7 +57,7 @@ jobs:
     if: github.ref_type == 'branch'
     needs: python-build
     secrets: inherit
-    uses: rapidsai/shared-action-workflows/.github/workflows/custom-job.yaml@branch-23.12
+    uses: rapidsai/shared-action-workflows/.github/workflows/custom-job.yaml@cuda-120-arm
     with:
       arch: "amd64"
       branch: ${{ inputs.branch }}
@@ -69,7 +69,7 @@ jobs:
       sha: ${{ inputs.sha }}
   wheel-build-cudf:
     secrets: inherit
-    uses: rapidsai/shared-action-workflows/.github/workflows/wheels-build.yaml@branch-23.12
+    uses: rapidsai/shared-action-workflows/.github/workflows/wheels-build.yaml@cuda-120-arm
     with:
       build_type: ${{ inputs.build_type || 'branch' }}
       branch: ${{ inputs.branch }}
@@ -79,7 +79,7 @@ jobs:
   wheel-publish-cudf:
     needs: wheel-build-cudf
     secrets: inherit
-    uses: rapidsai/shared-action-workflows/.github/workflows/wheels-publish.yaml@branch-23.12
+    uses: rapidsai/shared-action-workflows/.github/workflows/wheels-publish.yaml@cuda-120-arm
     with:
       build_type: ${{ inputs.build_type || 'branch' }}
       branch: ${{ inputs.branch }}
@@ -89,7 +89,7 @@ jobs:
   wheel-build-dask-cudf:
     needs: wheel-publish-cudf
     secrets: inherit
-    uses: rapidsai/shared-action-workflows/.github/workflows/wheels-build.yaml@branch-23.12
+    uses: rapidsai/shared-action-workflows/.github/workflows/wheels-build.yaml@cuda-120-arm
     with:
       matrix_filter: map(select(.ARCH == "amd64" and .PY_VER == "3.10" and (.CUDA_VER == "11.8.0" or .CUDA_VER == "12.0.1")))
       build_type: ${{ inputs.build_type || 'branch' }}
@@ -100,7 +100,7 @@ jobs:
   wheel-publish-dask-cudf:
     needs: wheel-build-dask-cudf
     secrets: inherit
-    uses: rapidsai/shared-action-workflows/.github/workflows/wheels-publish.yaml@branch-23.12
+    uses: rapidsai/shared-action-workflows/.github/workflows/wheels-publish.yaml@cuda-120-arm
     with:
       build_type: ${{ inputs.build_type || 'branch' }}
       branch: ${{ inputs.branch }}

diff --git a/.github/workflows/pr.yaml b/.github/workflows/pr.yaml
@@ -26,49 +26,49 @@ jobs:
       - wheel-build-dask-cudf
       - wheel-tests-dask-cudf
     secrets: inherit
-    uses: rapidsai/shared-action-workflows/.github/workflows/pr-builder.yaml@branch-23.12
+    uses: rapidsai/shared-action-workflows/.github/workflows/pr-builder.yaml@cuda-120-arm
   checks:
     secrets: inherit
-    uses: rapidsai/shared-action-workflows/.github/workflows/checks.yaml@branch-23.12
+    uses: rapidsai/shared-action-workflows/.github/workflows/checks.yaml@cuda-120-arm
     with:
       enable_check_generated_files: false
   conda-cpp-build:
     needs: checks
     secrets: inherit
-    uses: rapidsai/shared-action-workflows/.github/workflows/conda-cpp-build.yaml@branch-23.12
+    uses: rapidsai/shared-action-workflows/.github/workflows/conda-cpp-build.yaml@cuda-120-arm
     with:
       build_type: pull-request
   conda-cpp-tests:
     needs: conda-cpp-build
     secrets: inherit
-    uses: rapidsai/shared-action-workflows/.github/workflows/conda-cpp-tests.yaml@branch-23.12
+    uses: rapidsai/shared-action-workflows/.github/workflows/conda-cpp-tests.yaml@cuda-120-arm
     with:
       build_type: pull-request
   conda-python-build:
     needs: conda-cpp-build
     secrets: inherit
-    uses: rapidsai/shared-action-workflows/.github/workflows/conda-python-build.yaml@branch-23.12
+    uses: rapidsai/shared-action-workflows/.github/workflows/conda-python-build.yaml@cuda-120-arm
     with:
       build_type: pull-request
   conda-python-cudf-tests:
     needs: conda-python-build
     secrets: inherit
-    uses: rapidsai/shared-action-workflows/.github/workflows/conda-python-tests.yaml@branch-23.12
+    uses: rapidsai/shared-action-workflows/.github/workflows/conda-python-tests.yaml@cuda-120-arm
     with:
       build_type: pull-request
       test_script: "ci/test_python_cudf.sh"
   conda-python-other-tests:
     # Tests for dask_cudf, custreamz, cudf_kafka are separated for CI parallelism
     needs: conda-python-build
     secrets: inherit
-    uses: rapidsai/shared-action-workflows/.github/workflows/conda-python-tests.yaml@branch-23.12
+    uses: rapidsai/shared-action-workflows/.github/workflows/conda-python-tests.yaml@cuda-120-arm
     with:
       build_type: pull-request
       test_script: "ci/test_python_other.sh"
   conda-java-tests:
     needs: conda-cpp-build
     secrets: inherit
-    uses: rapidsai/shared-action-workflows/.github/workflows/custom-job.yaml@branch-23.12
+    uses: rapidsai/shared-action-workflows/.github/workflows/custom-job.yaml@cuda-120-arm
     with:
       build_type: pull-request
       node_type: "gpu-v100-latest-1"
@@ -78,7 +78,7 @@ jobs:
   conda-notebook-tests:
     needs: conda-python-build
     secrets: inherit
-    uses: rapidsai/shared-action-workflows/.github/workflows/custom-job.yaml@branch-23.12
+    uses: rapidsai/shared-action-workflows/.github/workflows/custom-job.yaml@cuda-120-arm
     with:
       build_type: pull-request
       node_type: "gpu-v100-latest-1"
@@ -88,7 +88,7 @@ jobs:
   docs-build:
     needs: conda-python-build
     secrets: inherit
-    uses: rapidsai/shared-action-workflows/.github/workflows/custom-job.yaml@branch-23.12
+    uses: rapidsai/shared-action-workflows/.github/workflows/custom-job.yaml@cuda-120-arm
     with:
       build_type: pull-request
       node_type: "gpu-v100-latest-1"
@@ -98,29 +98,29 @@ jobs:
   wheel-build-cudf:
     needs: checks
     secrets: inherit
-    uses: rapidsai/shared-action-workflows/.github/workflows/wheels-build.yaml@branch-23.12
+    uses: rapidsai/shared-action-workflows/.github/workflows/wheels-build.yaml@cuda-120-arm
     with:
       build_type: pull-request
       script: "ci/build_wheel_cudf.sh"
   wheel-tests-cudf:
     needs: wheel-build-cudf
     secrets: inherit
-    uses: rapidsai/shared-action-workflows/.github/workflows/wheels-test.yaml@branch-23.12
+    uses: rapidsai/shared-action-workflows/.github/workflows/wheels-test.yaml@cuda-120-arm
     with:
       build_type: pull-request
       script: ci/test_wheel_cudf.sh
   wheel-build-dask-cudf:
     needs: wheel-tests-cudf
     secrets: inherit
-    uses: rapidsai/shared-action-workflows/.github/workflows/wheels-build.yaml@branch-23.12
+    uses: rapidsai/shared-action-workflows/.github/workflows/wheels-build.yaml@cuda-120-arm
     with:
       matrix_filter: map(select(.ARCH == "amd64" and .PY_VER == "3.10" and (.CUDA_VER == "11.8.0" or .CUDA_VER == "12.0.1")))
       build_type: pull-request
       script: "ci/build_wheel_dask_cudf.sh"
   wheel-tests-dask-cudf:
     needs: wheel-build-dask-cudf
     secrets: inherit
-    uses: rapidsai/shared-action-workflows/.github/workflows/wheels-test.yaml@branch-23.12
+    uses: rapidsai/shared-action-workflows/.github/workflows/wheels-test.yaml@cuda-120-arm
     with:
       matrix_filter: map(select(.ARCH == "amd64" and .PY_VER == "3.10" and (.CUDA_VER == "11.8.0" or .CUDA_VER == "12.0.1")))
       build_type: pull-request

diff --git a/.github/workflows/test.yaml b/.github/workflows/test.yaml
@@ -16,15 +16,15 @@ on:
 jobs:
   conda-cpp-tests:
     secrets: inherit
-    uses: rapidsai/shared-action-workflows/.github/workflows/conda-cpp-tests.yaml@branch-23.12
+    uses: rapidsai/shared-action-workflows/.github/workflows/conda-cpp-tests.yaml@cuda-120-arm
     with:
       build_type: nightly
       branch: ${{ inputs.branch }}
       date: ${{ inputs.date }}
       sha: ${{ inputs.sha }}
   conda-cpp-memcheck-tests:
     secrets: inherit
-    uses: rapidsai/shared-action-workflows/.github/workflows/custom-job.yaml@branch-23.12
+    uses: rapidsai/shared-action-workflows/.github/workflows/custom-job.yaml@cuda-120-arm
     with:
       build_type: nightly
       branch: ${{ inputs.branch }}
@@ -36,7 +36,7 @@ jobs:
       run_script: "ci/test_cpp_memcheck.sh"
   conda-python-cudf-tests:
     secrets: inherit
-    uses: rapidsai/shared-action-workflows/.github/workflows/conda-python-tests.yaml@branch-23.12
+    uses: rapidsai/shared-action-workflows/.github/workflows/conda-python-tests.yaml@cuda-120-arm
     with:
       build_type: nightly
       branch: ${{ inputs.branch }}
@@ -46,7 +46,7 @@ jobs:
   conda-python-other-tests:
     # Tests for dask_cudf, custreamz, cudf_kafka are separated for CI parallelism
     secrets: inherit
-    uses: rapidsai/shared-action-workflows/.github/workflows/conda-python-tests.yaml@branch-23.12
+    uses: rapidsai/shared-action-workflows/.github/workflows/conda-python-tests.yaml@cuda-120-arm
     with:
       build_type: nightly
       branch: ${{ inputs.branch }}
@@ -55,7 +55,7 @@ jobs:
       test_script: "ci/test_python_other.sh"
   conda-java-tests:
     secrets: inherit
-    uses: rapidsai/shared-action-workflows/.github/workflows/custom-job.yaml@branch-23.12
+    uses: rapidsai/shared-action-workflows/.github/workflows/custom-job.yaml@cuda-120-arm
     with:
       build_type: nightly
       branch: ${{ inputs.branch }}
@@ -67,7 +67,7 @@ jobs:
       run_script: "ci/test_java.sh"
   conda-notebook-tests:
     secrets: inherit
-    uses: rapidsai/shared-action-workflows/.github/workflows/custom-job.yaml@branch-23.12
+    uses: rapidsai/shared-action-workflows/.github/workflows/custom-job.yaml@cuda-120-arm
     with:
       build_type: nightly
       branch: ${{ inputs.branch }}
@@ -79,7 +79,7 @@ jobs:
       run_script: "ci/test_notebooks.sh"
   wheel-tests-cudf:
     secrets: inherit
-    uses: rapidsai/shared-action-workflows/.github/workflows/wheels-test.yaml@branch-23.12
+    uses: rapidsai/shared-action-workflows/.github/workflows/wheels-test.yaml@cuda-120-arm
     with:
       build_type: nightly
       branch: ${{ inputs.branch }}
@@ -88,7 +88,7 @@ jobs:
       script: ci/test_wheel_cudf.sh
   wheel-tests-dask-cudf:
     secrets: inherit
-    uses: rapidsai/shared-action-workflows/.github/workflows/wheels-test.yaml@branch-23.12
+    uses: rapidsai/shared-action-workflows/.github/workflows/wheels-test.yaml@cuda-120-arm
     with:
       matrix_filter: map(select(.ARCH == "amd64" and .PY_VER == "3.10" and (.CUDA_VER == "11.8.0" or .CUDA_VER == "12.0.1")))
       build_type: nightly

diff --git a/cpp/include/cudf/binaryop.hpp b/cpp/include/cudf/binaryop.hpp
@@ -102,6 +102,7 @@ enum class binary_operator : int32_t {
  * @param rhs         The right operand column
  * @param op          The binary operator
  * @param output_type The desired data type of the output column
+ * @param stream CUDA stream used for device memory operations and kernel launches
  * @param mr          Device memory resource used to allocate the returned column's device memory
  * @return            Output column of `output_type` type containing the result of
  *                    the binary operation
@@ -115,6 +116,7 @@ std::unique_ptr<column> binary_operation(
   column_view const& rhs,
   binary_operator op,
   data_type output_type,
+  rmm::cuda_stream_view stream        = cudf::get_default_stream(),
   rmm::mr::device_memory_resource* mr = rmm::mr::get_current_device_resource());
 
 /**
@@ -131,6 +133,7 @@ std::unique_ptr<column> binary_operation(
  * @param rhs         The right operand scalar
  * @param op          The binary operator
  * @param output_type The desired data type of the output column
+ * @param stream CUDA stream used for device memory operations and kernel launches
  * @param mr          Device memory resource used to allocate the returned column's device memory
  * @return            Output column of `output_type` type containing the result of
  *                    the binary operation
@@ -144,6 +147,7 @@ std::unique_ptr<column> binary_operation(
   scalar const& rhs,
   binary_operator op,
   data_type output_type,
+  rmm::cuda_stream_view stream        = cudf::get_default_stream(),
   rmm::mr::device_memory_resource* mr = rmm::mr::get_current_device_resource());
 
 /**
@@ -158,6 +162,7 @@ std::unique_ptr<column> binary_operation(
  * @param rhs         The right operand column
  * @param op          The binary operator
  * @param output_type The desired data type of the output column
+ * @param stream CUDA stream used for device memory operations and kernel launches
  * @param mr          Device memory resource used to allocate the returned column's device memory
  * @return            Output column of `output_type` type containing the result of
  *                    the binary operation
@@ -172,6 +177,7 @@ std::unique_ptr<column> binary_operation(
   column_view const& rhs,
   binary_operator op,
   data_type output_type,
+  rmm::cuda_stream_view stream        = cudf::get_default_stream(),
   rmm::mr::device_memory_resource* mr = rmm::mr::get_current_device_resource());
 
 /**
@@ -189,6 +195,7 @@ std::unique_ptr<column> binary_operation(
  * @param output_type The desired data type of the output column. It is assumed
  *                    that output_type is compatible with the output data type
  *                    of the function in the PTX code
+ * @param stream CUDA stream used for device memory operations and kernel launches
  * @param mr          Device memory resource used to allocate the returned column's device memory
  * @return            Output column of `output_type` type containing the result of
  *                    the binary operation
@@ -201,6 +208,7 @@ std::unique_ptr<column> binary_operation(
   column_view const& rhs,
   std::string const& ptx,
   data_type output_type,
+  rmm::cuda_stream_view stream        = cudf::get_default_stream(),
   rmm::mr::device_memory_resource* mr = rmm::mr::get_current_device_resource());
 
 /**

diff --git a/cpp/include/cudf/detail/utilities/pinned_host_vector.hpp b/cpp/include/cudf/detail/utilities/pinned_host_vector.hpp
@@ -169,7 +169,12 @@ class pinned_allocator {
    *        It is the responsibility of the caller to destroy
    *        the objects stored at \p p.
    */
-  __host__ inline void deallocate(pointer p, size_type /*cnt*/) { CUDF_CUDA_TRY(cudaFreeHost(p)); }
+  __host__ inline void deallocate(pointer p, size_type /*cnt*/)
+  {
+    auto dealloc_worked = cudaFreeHost(p);
+    (void)dealloc_worked;
+    assert(dealloc_worked == cudaSuccess);
+  }
 
   /**
    * @brief This method returns the maximum size of the \c cnt parameter

diff --git a/cpp/include/cudf/io/detail/parquet.hpp b/cpp/include/cudf/io/detail/parquet.hpp
@@ -38,7 +38,7 @@ class parquet_reader_options;
 class parquet_writer_options;
 class chunked_parquet_writer_options;
 
-namespace detail::parquet {
+namespace parquet::detail {
 
 /**
  * @brief Class to read Parquet dataset data into columns.
@@ -186,7 +186,7 @@ class writer {
    */
   explicit writer(std::vector<std::unique_ptr<data_sink>> sinks,
                   parquet_writer_options const& options,
-                  single_write_mode mode,
+                  cudf::io::detail::single_write_mode mode,
                   rmm::cuda_stream_view stream);
 
   /**
@@ -201,7 +201,7 @@ class writer {
    */
   explicit writer(std::vector<std::unique_ptr<data_sink>> sinks,
                   chunked_parquet_writer_options const& options,
-                  single_write_mode mode,
+                  cudf::io::detail::single_write_mode mode,
                   rmm::cuda_stream_view stream);
 
   /**
@@ -250,5 +250,5 @@ class writer {
  * metadata.
  */
 parquet_metadata read_parquet_metadata(host_span<std::unique_ptr<datasource> const> sources);
-}  // namespace detail::parquet
+}  // namespace parquet::detail
 }  // namespace cudf::io
diff --git a/cpp/include/cudf/io/parquet.hpp b/cpp/include/cudf/io/parquet.hpp
@@ -499,7 +499,7 @@ class chunked_parquet_reader {
   [[nodiscard]] table_with_metadata read_chunk() const;
 
  private:
-  std::unique_ptr<cudf::io::detail::parquet::chunked_reader> reader;
+  std::unique_ptr<cudf::io::parquet::detail::chunked_reader> reader;
 };
 
 /** @} */  // end of group
@@ -1750,7 +1750,7 @@ class parquet_chunked_writer {
     std::vector<std::string> const& column_chunks_file_paths = {});
 
   /// Unique pointer to impl writer class
-  std::unique_ptr<cudf::io::detail::parquet::writer> writer;
+  std::unique_ptr<parquet::detail::writer> writer;
 };
 
 /** @} */  // end of group