Merge branch 'branch-25.02' into mukernels_strings

.github/workflows/pr.yaml

@@ -13,6 +13,7 @@ jobs:
   # Please keep pr-builder as the top job here
   pr-builder:
     needs:
+      - check-nightly-ci
       - changed-files
       - checks
       - conda-cpp-build
@@ -54,6 +55,18 @@ jobs:
       - name: Telemetry setup
         if: ${{ vars.TELEMETRY_ENABLED == 'true' }}
         uses: rapidsai/shared-actions/telemetry-dispatch-stash-base-env-vars@main
+  check-nightly-ci:
+    # Switch to ubuntu-latest once it defaults to a version of Ubuntu that
+    # provides at least Python 3.11 (see
+    # https://docs.python.org/3/library/datetime.html#datetime.date.fromisoformat)
+    runs-on: ubuntu-24.04
+    env:
+      RAPIDS_GH_TOKEN: ${{ secrets.GITHUB_TOKEN }}
+    steps:
+      - name: Check if nightly CI is passing
+        uses: rapidsai/shared-actions/check_nightly_success/dispatch@main
+        with:
+          repo: cudf
   changed-files:
     secrets: inherit
     needs: telemetry-setup
@@ -328,16 +341,11 @@ jobs:
         run_script: "ci/cudf_pandas_scripts/pandas-tests/diff.sh"
 
   telemetry-summarize:
-    runs-on: ubuntu-latest
+    # This job must use a self-hosted runner to record telemetry traces.
+    runs-on: linux-amd64-cpu4
     needs: pr-builder
     if: ${{ vars.TELEMETRY_ENABLED == 'true' && !cancelled() }}
     continue-on-error: true
     steps:
-      - name: Load stashed telemetry env vars
-        uses: rapidsai/shared-actions/telemetry-dispatch-load-base-env-vars@main
-        with:
-            load_service_name: true
       - name: Telemetry summarize
-        uses: rapidsai/shared-actions/telemetry-dispatch-write-summary@main
-        with:
-          cert_concat: "${{ secrets.OTEL_EXPORTER_OTLP_CA_CERTIFICATE }};${{ secrets.OTEL_EXPORTER_OTLP_CLIENT_CERTIFICATE }};${{ secrets.OTEL_EXPORTER_OTLP_CLIENT_KEY }}"
+        uses: rapidsai/shared-actions/telemetry-dispatch-summarize@main

conda/environments/all_cuda-118_arch-x86_64.yaml

@@ -71,7 +71,7 @@ dependencies:
 - ptxcompiler
 - pyarrow>=14.0.0,<19.0.0a0
 - pydata-sphinx-theme!=0.14.2
-- pynvml>=11.4.1,<12.0.0a0
+- pynvml>=12.0.0,<13.0.0a0
 - pytest-benchmark
 - pytest-cases>=3.8.2
 - pytest-cov

conda/environments/all_cuda-125_arch-x86_64.yaml

@@ -69,7 +69,7 @@ dependencies:
 - pyarrow>=14.0.0,<19.0.0a0
 - pydata-sphinx-theme!=0.14.2
 - pynvjitlink>=0.0.0a0
-- pynvml>=11.4.1,<12.0.0a0
+- pynvml>=12.0.0,<13.0.0a0
 - pytest-benchmark
 - pytest-cases>=3.8.2
 - pytest-cov

conda/recipes/dask-cudf/meta.yaml

@@ -43,7 +43,7 @@ requirements:
   run:
     - python
     - cudf ={{ version }}
-    - pynvml >=11.4.1,<12.0.0a0
+    - pynvml >=12.0.0,<13.0.0a0
     - rapids-dask-dependency ={{ minor_version }}
     - {{ pin_compatible('cuda-version', max_pin='x', min_pin='x') }}
 

cpp/CMakeLists.txt

@@ -1,5 +1,5 @@
 # =============================================================================
-# Copyright (c) 2018-2024, NVIDIA CORPORATION.
+# Copyright (c) 2018-2025, NVIDIA CORPORATION.
 #
 # Licensed under the Apache License, Version 2.0 (the "License"); you may not use this file except
 # in compliance with the License. You may obtain a copy of the License at
@@ -274,11 +274,8 @@ endif()
 # add third party dependencies using CPM
 rapids_cpm_init()
 
-# Not using rapids-cmake since we never want to find, always download.
-CPMAddPackage(
-  NAME rapids_logger GITHUB_REPOSITORY rapidsai/rapids-logger GIT_SHALLOW TRUE GIT_TAG
-  c510947ae9d3a67530cfe3e5eaccb5a3b8ea0e55 VERSION c510947ae9d3a67530cfe3e5eaccb5a3b8ea0e55
-)
+include(${rapids-cmake-dir}/cpm/rapids_logger.cmake)
+rapids_cpm_rapids_logger()
 rapids_make_logger(cudf EXPORT_SET cudf-exports)
 
 # find jitify
@@ -446,14 +443,15 @@ add_library(
   src/groupby/sort/group_quantiles.cu
   src/groupby/sort/group_std.cu
   src/groupby/sort/group_sum.cu
-  src/groupby/sort/scan.cpp
   src/groupby/sort/group_count_scan.cu
   src/groupby/sort/group_max_scan.cu
   src/groupby/sort/group_min_scan.cu
   src/groupby/sort/group_product_scan.cu
   src/groupby/sort/group_rank_scan.cu
   src/groupby/sort/group_replace_nulls.cu
   src/groupby/sort/group_sum_scan.cu
+  src/groupby/sort/host_udf_aggregation.cpp
+  src/groupby/sort/scan.cpp
   src/groupby/sort/sort_helper.cu
   src/hash/md5_hash.cu
   src/hash/murmurhash3_x86_32.cu
@@ -916,7 +914,9 @@ if(CUDF_LARGE_STRINGS_DISABLED)
 endif()
 
 # Define logging level
-target_compile_definitions(cudf PRIVATE "CUDF_LOG_ACTIVE_LEVEL=${LIBCUDF_LOGGING_LEVEL}")
+target_compile_definitions(
+  cudf PRIVATE "CUDF_LOG_ACTIVE_LEVEL=CUDF_LOG_LEVEL_${LIBCUDF_LOGGING_LEVEL}"
+)
 
 # Enable remote IO through KvikIO
 target_compile_definitions(cudf PRIVATE $<$<BOOL:${CUDF_KVIKIO_REMOTE_IO}>:CUDF_KVIKIO_REMOTE_IO>)

cpp/cmake/thirdparty/get_nanoarrow.cmake

@@ -14,11 +14,6 @@
 
 # This function finds nanoarrow and sets any additional necessary environment variables.
 function(find_and_configure_nanoarrow)
-  include(${rapids-cmake-dir}/cpm/package_override.cmake)
-
-  set(cudf_patch_dir "${CMAKE_CURRENT_FUNCTION_LIST_DIR}/patches")
-  rapids_cpm_package_override("${cudf_patch_dir}/nanoarrow_override.json")
-
   if(NOT BUILD_SHARED_LIBS)
     set(_exclude_from_all EXCLUDE_FROM_ALL FALSE)
   else()
@@ -31,6 +26,9 @@ function(find_and_configure_nanoarrow)
     nanoarrow 0.6.0.dev
     GLOBAL_TARGETS nanoarrow
     CPM_ARGS
+    GIT_REPOSITORY https://github.com/apache/arrow-nanoarrow.git
+    GIT_TAG 1e2664a70ec14907409cadcceb14d79b9670bcdb
+    GIT_SHALLOW FALSE
     OPTIONS "BUILD_SHARED_LIBS OFF" "NANOARROW_NAMESPACE cudf" ${_exclude_from_all}
   )
   set_target_properties(nanoarrow PROPERTIES POSITION_INDEPENDENT_CODE ON)

cpp/include/cudf/aggregation.hpp

@@ -110,8 +110,9 @@ class aggregation {
     COLLECT_SET,     ///< collect values into a list without duplicate entries
     LEAD,            ///< window function, accesses row at specified offset following current row
     LAG,             ///< window function, accesses row at specified offset preceding current row
-    PTX,             ///< PTX  UDF based reduction
-    CUDA,            ///< CUDA UDF based reduction
+    PTX,             ///< PTX  based UDF aggregation
+    CUDA,            ///< CUDA based UDF aggregation
+    HOST_UDF,        ///< host based UDF aggregation
     MERGE_LISTS,     ///< merge multiple lists values into one list
     MERGE_SETS,      ///< merge multiple lists values into one list then drop duplicate entries
     MERGE_M2,        ///< merge partial values of M2 aggregation,
@@ -120,7 +121,7 @@ class aggregation {
     TDIGEST,         ///< create a tdigest from a set of input values
     MERGE_TDIGEST,   ///< create a tdigest by merging multiple tdigests together
     HISTOGRAM,       ///< compute frequency of each element
-    MERGE_HISTOGRAM  ///< merge partial values of HISTOGRAM aggregation,
+    MERGE_HISTOGRAM  ///< merge partial values of HISTOGRAM aggregation
   };
 
   aggregation() = delete;
@@ -599,6 +600,18 @@ std::unique_ptr<Base> make_udf_aggregation(udf_type type,
                                            std::string const& user_defined_aggregator,
                                            data_type output_type);
 
+// Forward declaration of `host_udf_base` for the factory function of `HOST_UDF` aggregation.
+struct host_udf_base;
+
+/**
+ * @brief Factory to create a HOST_UDF aggregation.
+ *
+ * @param host_udf An instance of a class derived from `host_udf_base` to perform aggregation
+ * @return A HOST_UDF aggregation object
+ */
+template <typename Base = aggregation>
+std::unique_ptr<Base> make_host_udf_aggregation(std::unique_ptr<host_udf_base> host_udf);
+
 /**
  * @brief Factory to create a MERGE_LISTS aggregation.
  *