Merge pull request #2561 from bdice/branch-25.04-merge-25.02

Forward-merge branch-25.02 into branch-25.04

.devcontainer/cuda11.8-pip/devcontainer.json

@@ -5,7 +5,7 @@
     "args": {
       "CUDA": "11.8",
       "PYTHON_PACKAGE_MANAGER": "pip",
-      "BASE": "rapidsai/devcontainers:25.04-cpp-cuda11.8-ucx1.17.0-openmpi-ubuntu22.04"
+      "BASE": "rapidsai/devcontainers:25.04-cpp-cuda11.8-ucx1.18.0-openmpi-ubuntu22.04"
     }
   },
   "runArgs": [

.devcontainer/cuda12.5-conda/devcontainer.json → .devcontainer/cuda12.8-conda/devcontainer.json

@@ -3,15 +3,15 @@
     "context": "${localWorkspaceFolder}/.devcontainer",
     "dockerfile": "${localWorkspaceFolder}/.devcontainer/Dockerfile",
     "args": {
-      "CUDA": "12.5",
+      "CUDA": "12.8",
       "PYTHON_PACKAGE_MANAGER": "conda",
       "BASE": "rapidsai/devcontainers:25.04-cpp-mambaforge-ubuntu22.04"
     }
   },
   "runArgs": [
     "--rm",
     "--name",
-    "${localEnv:USER:anon}-rapids-${localWorkspaceFolderBasename}-25.04-cuda12.5-conda"
+    "${localEnv:USER:anon}-rapids-${localWorkspaceFolderBasename}-25.04-cuda12.8-conda"
   ],
   "hostRequirements": {"gpu": "optional"},
   "features": {
@@ -20,7 +20,7 @@
   "overrideFeatureInstallOrder": [
     "ghcr.io/rapidsai/devcontainers/features/rapids-build-utils"
   ],
-  "initializeCommand": ["/bin/bash", "-c", "mkdir -m 0755 -p ${localWorkspaceFolder}/../.{aws,cache,config,conda/pkgs,conda/${localWorkspaceFolderBasename}-cuda12.5-envs}"],
+  "initializeCommand": ["/bin/bash", "-c", "mkdir -m 0755 -p ${localWorkspaceFolder}/../.{aws,cache,config,conda/pkgs,conda/${localWorkspaceFolderBasename}-cuda12.8-envs}"],
   "postAttachCommand": ["/bin/bash", "-c", "if [ ${CODESPACES:-false} = 'true' ]; then . devcontainer-utils-post-attach-command; . rapids-post-attach-command; fi"],
   "workspaceFolder": "/home/coder",
   "workspaceMount": "source=${localWorkspaceFolder},target=/home/coder/raft,type=bind,consistency=consistent",
@@ -29,7 +29,7 @@
     "source=${localWorkspaceFolder}/../.cache,target=/home/coder/.cache,type=bind,consistency=consistent",
     "source=${localWorkspaceFolder}/../.config,target=/home/coder/.config,type=bind,consistency=consistent",
     "source=${localWorkspaceFolder}/../.conda/pkgs,target=/home/coder/.conda/pkgs,type=bind,consistency=consistent",
-    "source=${localWorkspaceFolder}/../.conda/${localWorkspaceFolderBasename}-cuda12.5-envs,target=/home/coder/.conda/envs,type=bind,consistency=consistent"
+    "source=${localWorkspaceFolder}/../.conda/${localWorkspaceFolderBasename}-cuda12.8-envs,target=/home/coder/.conda/envs,type=bind,consistency=consistent"
   ],
   "customizations": {
     "vscode": {

.devcontainer/cuda12.5-pip/devcontainer.json → .devcontainer/cuda12.8-pip/devcontainer.json

@@ -3,20 +3,20 @@
     "context": "${localWorkspaceFolder}/.devcontainer",
     "dockerfile": "${localWorkspaceFolder}/.devcontainer/Dockerfile",
     "args": {
-      "CUDA": "12.5",
+      "CUDA": "12.8",
       "PYTHON_PACKAGE_MANAGER": "pip",
-      "BASE": "rapidsai/devcontainers:25.04-cpp-cuda12.5-ucx1.17.0-openmpi-ubuntu22.04"
+      "BASE": "rapidsai/devcontainers:25.04-cpp-cuda12.8-ucx1.18.0-openmpi-ubuntu22.04"
     }
   },
   "runArgs": [
     "--rm",
     "--name",
-    "${localEnv:USER:anon}-rapids-${localWorkspaceFolderBasename}-25.04-cuda12.5-pip"
+    "${localEnv:USER:anon}-rapids-${localWorkspaceFolderBasename}-25.04-cuda12.8-pip"
   ],
   "hostRequirements": {"gpu": "optional"},
   "features": {
     "ghcr.io/rapidsai/devcontainers/features/cuda:25.4": {
-      "version": "12.5",
+      "version": "12.8",
       "installcuBLAS": true,
       "installcuSOLVER": true,
       "installcuRAND": true,
@@ -29,15 +29,15 @@
     "ghcr.io/rapidsai/devcontainers/features/cuda",
     "ghcr.io/rapidsai/devcontainers/features/rapids-build-utils"
   ],
-  "initializeCommand": ["/bin/bash", "-c", "mkdir -m 0755 -p ${localWorkspaceFolder}/../.{aws,cache,config/pip,local/share/${localWorkspaceFolderBasename}-cuda12.5-venvs}"],
+  "initializeCommand": ["/bin/bash", "-c", "mkdir -m 0755 -p ${localWorkspaceFolder}/../.{aws,cache,config/pip,local/share/${localWorkspaceFolderBasename}-cuda12.8-venvs}"],
   "postAttachCommand": ["/bin/bash", "-c", "if [ ${CODESPACES:-false} = 'true' ]; then . devcontainer-utils-post-attach-command; . rapids-post-attach-command; fi"],
   "workspaceFolder": "/home/coder",
   "workspaceMount": "source=${localWorkspaceFolder},target=/home/coder/raft,type=bind,consistency=consistent",
   "mounts": [
     "source=${localWorkspaceFolder}/../.aws,target=/home/coder/.aws,type=bind,consistency=consistent",
     "source=${localWorkspaceFolder}/../.cache,target=/home/coder/.cache,type=bind,consistency=consistent",
     "source=${localWorkspaceFolder}/../.config,target=/home/coder/.config,type=bind,consistency=consistent",
-    "source=${localWorkspaceFolder}/../.local/share/${localWorkspaceFolderBasename}-cuda12.5-venvs,target=/home/coder/.local/share/venvs,type=bind,consistency=consistent"
+    "source=${localWorkspaceFolder}/../.local/share/${localWorkspaceFolderBasename}-cuda12.8-venvs,target=/home/coder/.local/share/venvs,type=bind,consistency=consistent"
   ],
   "customizations": {
     "vscode": {

.github/workflows/pr.yaml

@@ -164,7 +164,7 @@ jobs:
     uses: rapidsai/shared-workflows/.github/workflows/build-in-devcontainer.yaml@branch-25.04
     with:
       arch: '["amd64"]'
-      cuda: '["12.5"]'
+      cuda: '["12.8"]'
       build_command: |
         sccache -z;
         build-all -DBUILD_PRIMS_BENCH=ON --verbose;

.pre-commit-config.yaml

@@ -1,6 +1,11 @@
-# Copyright (c) 2022-2023, NVIDIA CORPORATION.
+# Copyright (c) 2022-2025, NVIDIA CORPORATION.
 
 repos:
+      - repo: https://github.com/pre-commit/pre-commit-hooks
+        rev: v5.0.0
+        hooks:
+              - id: trailing-whitespace
+              - id: end-of-file-fixer
       - repo: https://github.com/PyCQA/isort
         rev: 5.12.0
         hooks:
@@ -98,7 +103,7 @@ repos:
                     ^CHANGELOG[.]md$|
                     ^cpp/cmake/patches/cutlass/build-export[.]patch$
       - repo: https://github.com/pre-commit/pre-commit-hooks
-        rev: v4.5.0
+        rev: v5.0.0
         hooks:
               - id: check-json
       - repo: https://github.com/rapidsai/pre-commit-hooks
@@ -118,7 +123,7 @@ repos:
                   docs/source/sphinxext/github_link[.]py|
           - id: verify-alpha-spec
       - repo: https://github.com/rapidsai/dependency-file-generator
-        rev: v1.16.0
+        rev: v1.17.0
         hooks:
             - id: rapids-dependency-file-generator
               args: ["--clean"]

README.md

@@ -240,15 +240,15 @@ mamba install -c rapidsai -c conda-forge -c nvidia raft-dask pylibraft cuda-vers
 
 ```bash
 # for CUDA 12.5
-mamba install -c rapidsai -c conda-forge -c nvidia raft-dask pylibraft cuda-version=12.5
+mamba install -c rapidsai -c conda-forge -c nvidia raft-dask pylibraft cuda-version=12.8
 ```
 
 Note that the above commands will also install `libraft-headers` and `libraft`.
 
 You can also install the conda packages individually using the `mamba` command above. For example, if you'd like to install RAFT's headers and pre-compiled shared library to use in your project:
 ```bash
 # for CUDA 12.5
-mamba install -c rapidsai -c conda-forge -c nvidia libraft libraft-headers cuda-version=12.5
+mamba install -c rapidsai -c conda-forge -c nvidia libraft libraft-headers cuda-version=12.8
 ```
 
 ### Installing Python through Pip

build.sh

@@ -473,4 +473,3 @@ if hasArg docs; then
     cd ${SPHINX_BUILD_DIR}
     sphinx-build -b html source _html
 fi
-

conda/environments/all_cuda-125_arch-aarch64.yaml → conda/environments/all_cuda-128_arch-aarch64.yaml

@@ -17,7 +17,7 @@ dependencies:
 - cuda-nvtx-dev
 - cuda-profiler-api
 - cuda-python>=12.6.2,<13.0a0
-- cuda-version=12.5
+- cuda-version=12.8
 - cupy>=12.0.0
 - cxx-compiler
 - cython>=3.0.0,<3.1.0a0
@@ -53,4 +53,4 @@ dependencies:
 - sphinx-markdown-tables
 - sysroot_linux-aarch64==2.28
 - ucx-py==0.43.*,>=0.0.0a0
-name: all_cuda-125_arch-aarch64
+name: all_cuda-128_arch-aarch64

conda/environments/all_cuda-125_arch-x86_64.yaml → conda/environments/all_cuda-128_arch-x86_64.yaml

@@ -17,7 +17,7 @@ dependencies:
 - cuda-nvtx-dev
 - cuda-profiler-api
 - cuda-python>=12.6.2,<13.0a0
-- cuda-version=12.5
+- cuda-version=12.8
 - cupy>=12.0.0
 - cxx-compiler
 - cython>=3.0.0,<3.1.0a0
@@ -53,4 +53,4 @@ dependencies:
 - sphinx-markdown-tables
 - sysroot_linux-64==2.28
 - ucx-py==0.43.*,>=0.0.0a0
-name: all_cuda-125_arch-x86_64
+name: all_cuda-128_arch-x86_64

cpp/cmake/patches/cutlass/build-export.patch

@@ -20,8 +20,7 @@ index 7419bdf5e..545384d82 100755
 -  $<BUILD_INTERFACE:${cute_SOURCE_DIR}/include>
 -  $<BUILD_INTERFACE:${cute_SOURCE_DIR}/examples>
    )
- 
+
  # Mark CTK headers as system to supress warnings from them
--- 
+--
 2.34.1
-

cpp/include/raft/cluster/detail/kmeans_auto_find_k.cuh

@@ -227,4 +227,4 @@ void find_k(raft::resources const& handle,
                                                               n_iter);
   }
 }
-}  // namespace raft::cluster::detail
+}  // namespace raft::cluster::detail

cpp/include/raft/cluster/detail/mst.cuh

@@ -204,4 +204,4 @@ void build_sorted_mst(
   raft::copy_async(mst_weight, mst_coo.weights.data(), mst_coo.n_edges, stream);
 }
 
-};  // namespace raft::cluster::detail
+};  // namespace raft::cluster::detail

cpp/include/raft/cluster/detail/single_linkage.cuh

@@ -122,4 +122,4 @@ void single_linkage(raft::resources const& handle,
   out->n_leaves               = m;
   out->n_connected_components = 1;
 }
-};  // namespace raft::cluster::detail
+};  // namespace raft::cluster::detail

cpp/include/raft/common/nvtx.hpp

@@ -21,4 +21,4 @@
 
 #pragma once
 
-#include <raft/core/nvtx.hpp>
+#include <raft/core/nvtx.hpp>

cpp/include/raft/core/coo_matrix.hpp

@@ -297,4 +297,4 @@ class coo_matrix
 
 /** @} */
 
-}  // namespace raft
+}  // namespace raft

cpp/include/raft/core/csr_matrix.hpp

@@ -309,4 +309,4 @@ class csr_matrix
 
 /** @} */
 
-}  // namespace raft
+}  // namespace raft

cpp/include/raft/core/detail/mdspan_util.cuh

@@ -67,4 +67,4 @@ MDSPAN_INLINE_FUNCTION auto popc(uint64_t v) -> int32_t
 #endif  // compiler
 }
 
-}  // end namespace raft::detail
+}  // end namespace raft::detail

cpp/include/raft/core/device_coo_matrix.hpp

@@ -395,4 +395,4 @@ auto make_device_coordinate_structure_view(raft::device_span<RowType> rows,
 
 /** @} */
 
-};  // namespace raft
+};  // namespace raft

cpp/include/raft/core/device_csr_matrix.hpp

@@ -422,4 +422,4 @@ auto make_device_compressed_structure_view(raft::device_span<IndptrType> indptr,
 
 /** @} */
 
-};  // namespace raft
+};  // namespace raft

cpp/include/raft/core/device_span.hpp

@@ -34,4 +34,4 @@ using device_span = span<T, true, extent>;
 /**
  * @}
  */
-}  // end namespace raft
+}  // end namespace raft

cpp/include/raft/core/host_coo_matrix.hpp

@@ -393,4 +393,4 @@ auto make_host_coordinate_structure_view(raft::host_span<RowType> rows,
 
 /** @} */
 
-};  // namespace raft
+};  // namespace raft

cpp/include/raft/core/host_csr_matrix.hpp

@@ -423,4 +423,4 @@ auto make_host_compressed_structure_view(raft::host_span<IndptrType> indptr,
 
 /** @} */
 
-};  // namespace raft
+};  // namespace raft

cpp/include/raft/core/host_mdarray.hpp

@@ -253,4 +253,4 @@ auto make_host_vector(IndexType n)
   return make_host_mdarray<ElementType, IndexType, LayoutPolicy>(make_extents<IndexType>(n));
 }
 
-}  // end namespace raft
+}  // end namespace raft

cpp/include/raft/core/host_span.hpp

@@ -35,4 +35,4 @@ using host_span = span<T, false, extent>;
  * @}
  */
 
-}  // end namespace raft
+}  // end namespace raft

cpp/include/raft/core/resource/device_id.hpp

@@ -73,4 +73,4 @@ inline int get_device_id(resources const& res)
 /**
  * @}
  */
-}  // namespace raft::resource
+}  // namespace raft::resource

cpp/include/raft/core/resource/device_properties.hpp

@@ -75,4 +75,4 @@ inline cudaDeviceProp& get_device_properties(resources const& res)
 /**
  * @}
  */
-}  // namespace raft::resource
+}  // namespace raft::resource

cpp/include/raft/core/resource/sub_comms.hpp

@@ -79,4 +79,4 @@ inline void set_subcomm(resources const& res,
  * @}
  */
 
-}  // namespace raft::resource
+}  // namespace raft::resource

cpp/include/raft/core/sparse_types.hpp

@@ -222,4 +222,4 @@ class sparse_matrix {
 
 /* @} */
 
-}  // namespace raft
+}  // namespace raft

cpp/include/raft/distance/detail/fused_distance_nn/gemm.h

@@ -406,4 +406,4 @@ struct FusedDistanceNNGemm<double,
 
 }  // namespace kernel
 }  // namespace gemm
-}  // namespace cutlass
+}  // namespace cutlass

cpp/include/raft/distance/detail/pairwise_distance_gemm.h

@@ -235,4 +235,4 @@ struct PairwiseDistanceGemm<double,
 
 }  // namespace kernel
 }  // namespace gemm
-}  // namespace cutlass
+}  // namespace cutlass

cpp/include/raft/distance/fused_distance_nn.cuh

@@ -15,4 +15,4 @@
  */
 #pragma once
 
-#include "fused_distance_nn-inl.cuh"
+#include "fused_distance_nn-inl.cuh"

cpp/include/raft/label/classlabels.cuh

@@ -118,4 +118,4 @@ void make_monotonic(Type* out, Type* in, size_t N, cudaStream_t stream, bool zer
 };  // namespace label
 };  // end namespace raft
 
-#endif
+#endif

cpp/include/raft/label/detail/merge_labels.cuh

@@ -155,4 +155,4 @@ void merge_labels(value_idx* labels_a,
 
 }  // namespace detail
 };  // namespace label
-};  // namespace raft
+};  // namespace raft

cpp/include/raft/label/merge_labels.cuh

@@ -68,4 +68,4 @@ void merge_labels(value_idx* labels_a,
 };  // namespace label
 };  // namespace raft
 
-#endif
+#endif

cpp/include/raft/linalg/cholesky_r1_update.cuh

@@ -139,4 +139,4 @@ void choleskyRank1Update(raft::resources const& handle,
 };  // namespace linalg
 };  // namespace raft
 
-#endif
+#endif

cpp/include/raft/linalg/coalesced_reduction.cuh

@@ -163,4 +163,4 @@ void coalesced_reduction(raft::resources const& handle,
 };  // end namespace linalg
 };  // end namespace raft
 
-#endif
+#endif

cpp/include/raft/linalg/detail/add.cuh

@@ -60,4 +60,4 @@ void addDevScalar(
 
 }  // namespace detail
 }  // namespace linalg
-}  // namespace raft
+}  // namespace raft

cpp/include/raft/linalg/divide.cuh

@@ -99,4 +99,4 @@ void divide_scalar(raft::resources const& handle,
 };  // end namespace linalg
 };  // end namespace raft
 
-#endif
+#endif

cpp/include/raft/linalg/eig.cuh

@@ -223,4 +223,4 @@ void eig_jacobi(raft::resources const& handle,
 };  // end namespace linalg
 };  // end namespace raft
 
-#endif
+#endif