Merge remote-tracking branch 'upstream/branch-23.10' into 14088

rapidsai · Sep 27, 2023 · b4d963e · b4d963e
2 parents ffce0f6 + 31e5670
commit b4d963e
Show file tree

Hide file tree

Showing 147 changed files with 5,848 additions and 2,782 deletions.
diff --git a/.github/ops-bot.yaml b/.github/ops-bot.yaml
@@ -5,4 +5,3 @@ auto_merger: true
 branch_checker: true
 label_checker: true
 release_drafter: true
-recently_updated: true
diff --git a/.github/workflows/add_to_project.yml b/.github/workflows/add_to_project.yml
diff --git a/.github/workflows/build.yaml b/.github/workflows/build.yaml
@@ -62,7 +62,7 @@ jobs:
       arch: "amd64"
       branch: ${{ inputs.branch }}
       build_type: ${{ inputs.build_type || 'branch' }}
-      container_image: "rapidsai/ci:latest"
+      container_image: "rapidsai/ci-conda:latest"
       date: ${{ inputs.date }}
       node_type: "gpu-v100-latest-1"
       run_script: "ci/build_docs.sh"

diff --git a/.github/workflows/new-issues-to-triage-projects.yml b/.github/workflows/new-issues-to-triage-projects.yml
diff --git a/.github/workflows/pr.yaml b/.github/workflows/pr.yaml
@@ -73,7 +73,7 @@ jobs:
       build_type: pull-request
       node_type: "gpu-v100-latest-1"
       arch: "amd64"
-      container_image: "rapidsai/ci:latest"
+      container_image: "rapidsai/ci-conda:latest"
       run_script: "ci/test_java.sh"
   conda-notebook-tests:
     needs: conda-python-build
@@ -83,7 +83,7 @@ jobs:
       build_type: pull-request
       node_type: "gpu-v100-latest-1"
       arch: "amd64"
-      container_image: "rapidsai/ci:latest"
+      container_image: "rapidsai/ci-conda:latest"
       run_script: "ci/test_notebooks.sh"
   docs-build:
     needs: conda-python-build
@@ -93,7 +93,7 @@ jobs:
       build_type: pull-request
       node_type: "gpu-v100-latest-1"
       arch: "amd64"
-      container_image: "rapidsai/ci:latest"
+      container_image: "rapidsai/ci-conda:latest"
       run_script: "ci/build_docs.sh"
   wheel-build-cudf:
     needs: checks

diff --git a/.github/workflows/test.yaml b/.github/workflows/test.yaml
@@ -32,7 +32,7 @@ jobs:
       sha: ${{ inputs.sha }}
       node_type: "gpu-v100-latest-1"
       arch: "amd64"
-      container_image: "rapidsai/ci:latest"
+      container_image: "rapidsai/ci-conda:latest"
       run_script: "ci/test_cpp_memcheck.sh"
   conda-python-cudf-tests:
     secrets: inherit
@@ -63,7 +63,7 @@ jobs:
       sha: ${{ inputs.sha }}
       node_type: "gpu-v100-latest-1"
       arch: "amd64"
-      container_image: "rapidsai/ci:latest"
+      container_image: "rapidsai/ci-conda:latest"
       run_script: "ci/test_java.sh"
   conda-notebook-tests:
     secrets: inherit
@@ -75,7 +75,7 @@ jobs:
       sha: ${{ inputs.sha }}
       node_type: "gpu-v100-latest-1"
       arch: "amd64"
-      container_image: "rapidsai/ci:latest"
+      container_image: "rapidsai/ci-conda:latest"
       run_script: "ci/test_notebooks.sh"
   wheel-tests-cudf:
     secrets: inherit

diff --git a/ci/build_docs.sh b/ci/build_docs.sh
@@ -30,7 +30,7 @@ export RAPIDS_DOCS_DIR="$(mktemp -d)"
 
 rapids-logger "Build CPP docs"
 pushd cpp/doxygen
-aws s3 cp s3://rapidsai-docs/librmm/${RAPIDS_VERSION_NUMBER}/html/rmm.tag . || echo "Failed to download rmm Doxygen tag"
+aws s3 cp s3://rapidsai-docs/librmm/html/${RAPIDS_VERSION_NUMBER}/rmm.tag . || echo "Failed to download rmm Doxygen tag"
 doxygen Doxyfile
 mkdir -p "${RAPIDS_DOCS_DIR}/libcudf/html"
 mv html/* "${RAPIDS_DOCS_DIR}/libcudf/html"

diff --git a/conda/environments/all_cuda-118_arch-x86_64.yaml b/conda/environments/all_cuda-118_arch-x86_64.yaml
@@ -9,6 +9,7 @@ channels:
 - nvidia
 dependencies:
 - aiobotocore>=2.2.0
+- aws-sdk-cpp<1.11
 - benchmark==1.8.0
 - boto3>=1.21.21
 - botocore>=1.24.21
@@ -55,8 +56,8 @@ dependencies:
 - nbsphinx
 - ninja
 - notebook
-- numba>=0.57
-- numpy>=1.21
+- numba>=0.57,<0.58
+- numpy>=1.21,<1.25
 - numpydoc
 - nvcc_linux-64=11.8
 - nvcomp==2.6.1

diff --git a/conda/environments/all_cuda-120_arch-x86_64.yaml b/conda/environments/all_cuda-120_arch-x86_64.yaml
@@ -9,6 +9,7 @@ channels:
 - nvidia
 dependencies:
 - aiobotocore>=2.2.0
+- aws-sdk-cpp<1.11
 - benchmark==1.8.0
 - boto3>=1.21.21
 - botocore>=1.24.21
@@ -54,8 +55,8 @@ dependencies:
 - nbsphinx
 - ninja
 - notebook
-- numba>=0.57
-- numpy>=1.21
+- numba>=0.57,<0.58
+- numpy>=1.21,<1.25
 - numpydoc
 - nvcomp==2.6.1
 - nvtx>=0.2.1

diff --git a/conda/recipes/cudf/meta.yaml b/conda/recipes/cudf/meta.yaml
@@ -78,8 +78,10 @@ requirements:
     - typing_extensions >=4.0.0
     - pandas >=1.3,<1.6.0dev0
     - cupy >=12.0.0
-    - numba >=0.57
-    - numpy >=1.21
+    # TODO: Pin to numba<0.58 until #14160 is resolved
+    - numba >=0.57,<0.58
+    # TODO: Pin to numpy<1.25 until cudf requires pandas 2
+    - numpy >=1.21,<1.25
     - {{ pin_compatible('pyarrow', max_pin='x.x.x') }}
     - libcudf ={{ version }}
     - {{ pin_compatible('rmm', max_pin='x.x') }}

diff --git a/conda/recipes/libcudf/conda_build_config.yaml b/conda/recipes/libcudf/conda_build_config.yaml
@@ -22,6 +22,9 @@ gbench_version:
 gtest_version:
   - ">=1.13.0"
 
+aws_sdk_cpp_version:
+  - "<1.11"
+
 libarrow_version:
   - "=12"
 

diff --git a/conda/recipes/libcudf/meta.yaml b/conda/recipes/libcudf/meta.yaml
@@ -74,6 +74,7 @@ requirements:
     - gtest {{ gtest_version }}
     - gmock {{ gtest_version }}
     - zlib {{ zlib_version }}
+    - aws-sdk-cpp {{ aws_sdk_cpp_version }}
 
 outputs:
   - name: libcudf
@@ -107,6 +108,7 @@ outputs:
         - dlpack {{ dlpack_version }}
         - gtest {{ gtest_version }}
         - gmock {{ gtest_version }}
+        - aws-sdk-cpp {{ aws_sdk_cpp_version }}
     test:
       commands:
         - test -f $PREFIX/lib/libcudf.so

diff --git a/cpp/CMakeLists.txt b/cpp/CMakeLists.txt
@@ -413,11 +413,13 @@ add_library(
   src/io/utilities/arrow_io_source.cpp
   src/io/utilities/column_buffer.cpp
   src/io/utilities/config_utils.cpp
+  src/io/utilities/data_casting.cu
   src/io/utilities/data_sink.cpp
   src/io/utilities/datasource.cpp
   src/io/utilities/file_io_utilities.cpp
   src/io/utilities/parsing_utils.cu
   src/io/utilities/row_selection.cpp
+  src/io/utilities/type_inference.cu
   src/io/utilities/trie.cu
   src/jit/cache.cpp
   src/jit/parser.cpp
@@ -616,6 +618,7 @@ add_library(
   src/text/subword/subword_tokenize.cu
   src/text/subword/wordpiece_tokenizer.cu
   src/text/tokenize.cu
+  src/text/vocabulary_tokenize.cu
   src/transform/bools_to_mask.cu
   src/transform/compute_column.cu
   src/transform/encode.cu

diff --git a/cpp/benchmarks/CMakeLists.txt b/cpp/benchmarks/CMakeLists.txt
@@ -173,7 +173,7 @@ ConfigureBench(ITERATOR_BENCH iterator/iterator.cu)
 # ##################################################################################################
 # * search benchmark ------------------------------------------------------------------------------
 ConfigureBench(SEARCH_BENCH search/search.cpp)
-ConfigureNVBench(SEARCH_NVBENCH search/contains.cpp)
+ConfigureNVBench(SEARCH_NVBENCH search/contains_scalar.cpp search/contains_table.cpp)
 
 # ##################################################################################################
 # * sort benchmark --------------------------------------------------------------------------------

diff --git a/cpp/benchmarks/io/cuio_common.cpp b/cpp/benchmarks/io/cuio_common.cpp
@@ -15,6 +15,7 @@
  */
 
 #include <benchmarks/io/cuio_common.hpp>
+#include <cudf/detail/utilities/integer_utils.hpp>
 #include <cudf/detail/utilities/logger.hpp>
 
 #include <cstdio>
@@ -141,17 +142,18 @@ std::vector<std::string> select_column_names(std::vector<std::string> const& col
   return col_names_to_read;
 }
 
-std::vector<cudf::size_type> segments_in_chunk(int num_segments, int num_chunks, int chunk)
+std::vector<cudf::size_type> segments_in_chunk(int num_segments, int num_chunks, int chunk_idx)
 {
   CUDF_EXPECTS(num_segments >= num_chunks,
                "Number of chunks cannot be greater than the number of segments in the file");
-  auto start_segment = [num_segments, num_chunks](int chunk) {
-    return num_segments * chunk / num_chunks;
-  };
-  std::vector<cudf::size_type> selected_segments;
-  for (auto segment = start_segment(chunk); segment < start_segment(chunk + 1); ++segment) {
-    selected_segments.push_back(segment);
-  }
+  CUDF_EXPECTS(chunk_idx < num_chunks,
+               "Chunk index must be smaller than the number of chunks in the file");
+
+  auto const segments_in_chunk = cudf::util::div_rounding_up_unsafe(num_segments, num_chunks);
+  auto const begin_segment     = std::min(chunk_idx * segments_in_chunk, num_segments);
+  auto const end_segment       = std::min(begin_segment + segments_in_chunk, num_segments);
+  std::vector<cudf::size_type> selected_segments(end_segment - begin_segment);
+  std::iota(selected_segments.begin(), selected_segments.end(), begin_segment);
 
   return selected_segments;
 }

diff --git a/cpp/benchmarks/io/orc/orc_reader_options.cpp b/cpp/benchmarks/io/orc/orc_reader_options.cpp
@@ -19,6 +19,7 @@
 #include <benchmarks/io/cuio_common.hpp>
 #include <benchmarks/io/nvbench_helpers.hpp>
 
+#include <cudf/detail/utilities/integer_utils.hpp>
 #include <cudf/io/orc.hpp>
 #include <cudf/io/orc_metadata.hpp>
 #include <cudf/utilities/default_stream.hpp>
@@ -30,7 +31,7 @@
 constexpr int64_t data_size = 512 << 20;
 // The number of separate read calls to use when reading files in multiple chunks
 // Each call reads roughly equal amounts of data
-constexpr int32_t chunked_read_num_chunks = 8;
+constexpr int32_t chunked_read_num_chunks = 4;
 
 std::vector<std::string> get_top_level_col_names(cudf::io::source_info const& source)
 {
@@ -88,7 +89,7 @@ void BM_orc_read_varying_options(nvbench::state& state,
 
   auto const num_stripes =
     cudf::io::read_orc_metadata(source_sink.make_source_info()).num_stripes();
-  cudf::size_type const chunk_row_cnt = view.num_rows() / num_chunks;
+  auto const chunk_row_cnt = cudf::util::div_rounding_up_unsafe(view.num_rows(), num_chunks);
 
   auto mem_stats_logger = cudf::memory_stats_logger();
   state.set_cuda_stream(nvbench::make_cuda_stream_view(cudf::get_default_stream().value()));
@@ -99,7 +100,6 @@ void BM_orc_read_varying_options(nvbench::state& state,
       timer.start();
       cudf::size_type rows_read = 0;
       for (int32_t chunk = 0; chunk < num_chunks; ++chunk) {
-        auto const is_last_chunk = chunk == (num_chunks - 1);
         switch (RowSelection) {
           case row_selection::ALL: break;
           case row_selection::STRIPES:
@@ -108,7 +108,6 @@ void BM_orc_read_varying_options(nvbench::state& state,
           case row_selection::NROWS:
             read_options.set_skip_rows(chunk * chunk_row_cnt);
             read_options.set_num_rows(chunk_row_cnt);
-            if (is_last_chunk) read_options.set_num_rows(-1);
             break;
           default: CUDF_FAIL("Unsupported row selection method");
         }
@@ -132,9 +131,6 @@ using col_selections = nvbench::enum_type_list<column_selection::ALL,
                                                column_selection::ALTERNATE,
                                                column_selection::FIRST_HALF,
                                                column_selection::SECOND_HALF>;
-using row_selections =
-  nvbench::enum_type_list<row_selection::ALL, row_selection::STRIPES, row_selection::NROWS>;
-
 NVBENCH_BENCH_TYPES(BM_orc_read_varying_options,
                     NVBENCH_TYPE_AXES(col_selections,
                                       nvbench::enum_type_list<row_selection::ALL>,
@@ -146,6 +142,8 @@ NVBENCH_BENCH_TYPES(BM_orc_read_varying_options,
     {"column_selection", "row_selection", "uses_index", "uses_numpy_dtype", "timestamp_type"})
   .set_min_samples(4);
 
+using row_selections =
+  nvbench::enum_type_list<row_selection::ALL, row_selection::NROWS, row_selection::STRIPES>;
 NVBENCH_BENCH_TYPES(BM_orc_read_varying_options,
                     NVBENCH_TYPE_AXES(nvbench::enum_type_list<column_selection::ALL>,
                                       row_selections,