Merge branch 'branch-25.02' into go

.github/workflows/pr.yaml

@@ -12,6 +12,7 @@ concurrency:
 jobs:
   pr-builder:
     needs:
+      - check-nightly-ci
       - changed-files
       - checks
       - conda-cpp-build
@@ -30,6 +31,18 @@ jobs:
     if: always()
     with:
       needs: ${{ toJSON(needs) }}
+  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: cuvs
   changed-files:
     secrets: inherit
     uses: rapidsai/shared-workflows/.github/workflows/[email protected]

.pre-commit-config.yaml

@@ -108,8 +108,7 @@ repos:
                   [.](cmake|cpp|cu|cuh|h|hpp|sh|pxd|py|pyx|rs)$|
                   CMakeLists[.]txt$|
                   CMakeLists_standalone[.]txt$|
-                  meta[.]yaml$|
-                  setup[.]cfg$
+                  meta[.]yaml$
             exclude: |
               (?x)
                   docs/source/sphinxext/github_link\.py|

CHANGELOG.md

@@ -1,3 +1,67 @@
+# cuvs 24.12.00 (11 Dec 2024)
+
+## 🚨 Breaking Changes
+
+- HNSW CPU Hierarchy ([#465](https://github.com/rapidsai/cuvs/pull/465)) [@divyegala](https://github.com/divyegala)
+- Use dashes in cuvs-bench package name. ([#417](https://github.com/rapidsai/cuvs/pull/417)) [@bdice](https://github.com/bdice)
+
+## 🐛 Bug Fixes
+
+- Skip IVF-PQ packing test for lists with not enough data ([#512](https://github.com/rapidsai/cuvs/pull/512)) [@achirkin](https://github.com/achirkin)
+- [BUG] Fix CAGRA filter ([#489](https://github.com/rapidsai/cuvs/pull/489)) [@enp1s0](https://github.com/enp1s0)
+- Add `kIsSingleSource` to `PairwiseDistanceEpilogueElementwise` ([#485](https://github.com/rapidsai/cuvs/pull/485)) [@KyleFromNVIDIA](https://github.com/KyleFromNVIDIA)
+- Fix include errors, header, and unsafe locks in iface.hpp ([#467](https://github.com/rapidsai/cuvs/pull/467)) [@achirkin](https://github.com/achirkin)
+- Fix an OOB error in device-side cuvs::neighbors::refine and CAGRA kern_prune ([#460](https://github.com/rapidsai/cuvs/pull/460)) [@achirkin](https://github.com/achirkin)
+- Put a ceiling on cuda-python ([#445](https://github.com/rapidsai/cuvs/pull/445)) [@bdice](https://github.com/bdice)
+- Enable NVTX in cuvs-cagra-search component ([#439](https://github.com/rapidsai/cuvs/pull/439)) [@achirkin](https://github.com/achirkin)
+- BUG: CAGRA multi-cta illegal access with bad queries ([#438](https://github.com/rapidsai/cuvs/pull/438)) [@achirkin](https://github.com/achirkin)
+- Fix index overflow in edge cases of CAGRA graph optimize ([#435](https://github.com/rapidsai/cuvs/pull/435)) [@achirkin](https://github.com/achirkin)
+- Fix correct call to brute force in generate groundtruth of cuvs-bench ([#427](https://github.com/rapidsai/cuvs/pull/427)) [@dantegd](https://github.com/dantegd)
+- Use Python for sccache hit rate computation. ([#420](https://github.com/rapidsai/cuvs/pull/420)) [@bdice](https://github.com/bdice)
+- Add `click` package to `cuvs-bench` conda recipe ([#408](https://github.com/rapidsai/cuvs/pull/408)) [@divyegala](https://github.com/divyegala)
+- Fix NVTX annotations ([#400](https://github.com/rapidsai/cuvs/pull/400)) [@achirkin](https://github.com/achirkin)
+
+## 📖 Documentation
+
+- [Doc] Fix CAGRA search sample code ([#484](https://github.com/rapidsai/cuvs/pull/484)) [@enp1s0](https://github.com/enp1s0)
+- Fix broken link in README.md references ([#473](https://github.com/rapidsai/cuvs/pull/473)) [@Azurethi](https://github.com/Azurethi)
+- Adding tech stack to docs ([#448](https://github.com/rapidsai/cuvs/pull/448)) [@cjnolet](https://github.com/cjnolet)
+- Fix Question Retrieval notebook ([#352](https://github.com/rapidsai/cuvs/pull/352)) [@lowener](https://github.com/lowener)
+
+## 🚀 New Features
+
+- Add C++ API scalar quantization ([#494](https://github.com/rapidsai/cuvs/pull/494)) [@mfoerste4](https://github.com/mfoerste4)
+- HNSW CPU Hierarchy ([#465](https://github.com/rapidsai/cuvs/pull/465)) [@divyegala](https://github.com/divyegala)
+- Add serialization API to brute-force ([#461](https://github.com/rapidsai/cuvs/pull/461)) [@lowener](https://github.com/lowener)
+- Add Question Retrieval notebook using Milvus ([#451](https://github.com/rapidsai/cuvs/pull/451)) [@lowener](https://github.com/lowener)
+- Migrate feature diff for NN Descent from RAFT to cuVS ([#421](https://github.com/rapidsai/cuvs/pull/421)) [@divyegala](https://github.com/divyegala)
+- Add --no-lap-sync cmd option to ann-bench ([#405](https://github.com/rapidsai/cuvs/pull/405)) [@achirkin](https://github.com/achirkin)
+- Add `InnerProduct` and `CosineExpanded` metric support in NN Descent ([#177](https://github.com/rapidsai/cuvs/pull/177)) [@divyegala](https://github.com/divyegala)
+
+## 🛠️ Improvements
+
+- Update cuvs to match raft's cutlass changes ([#516](https://github.com/rapidsai/cuvs/pull/516)) [@vyasr](https://github.com/vyasr)
+- add a README for wheels ([#504](https://github.com/rapidsai/cuvs/pull/504)) [@jameslamb](https://github.com/jameslamb)
+- Move check_input_array from pylibraft ([#474](https://github.com/rapidsai/cuvs/pull/474)) [@benfred](https://github.com/benfred)
+- use different wheel-size thresholds based on CUDA version ([#469](https://github.com/rapidsai/cuvs/pull/469)) [@jameslamb](https://github.com/jameslamb)
+- Modify cuvs-bench to be able to generate ground truth in CPU systems ([#466](https://github.com/rapidsai/cuvs/pull/466)) [@dantegd](https://github.com/dantegd)
+- enforce wheel size limits, README formatting in CI ([#464](https://github.com/rapidsai/cuvs/pull/464)) [@jameslamb](https://github.com/jameslamb)
+- Moving spectral embedding and kernel gramm APIs to cuVS ([#463](https://github.com/rapidsai/cuvs/pull/463)) [@cjnolet](https://github.com/cjnolet)
+- Migrate sparse knn and distances code from raft ([#457](https://github.com/rapidsai/cuvs/pull/457)) [@benfred](https://github.com/benfred)
+- Don't presume pointers location infers usability. ([#441](https://github.com/rapidsai/cuvs/pull/441)) [@robertmaynard](https://github.com/robertmaynard)
+- call `enable_testing` in root CMakeLists.txt ([#437](https://github.com/rapidsai/cuvs/pull/437)) [@robertmaynard](https://github.com/robertmaynard)
+- CAGRA tech debt: distance descriptor and workspace memory ([#436](https://github.com/rapidsai/cuvs/pull/436)) [@achirkin](https://github.com/achirkin)
+- Add ci run_ scripts needed for build infra ([#434](https://github.com/rapidsai/cuvs/pull/434)) [@robertmaynard](https://github.com/robertmaynard)
+- Use environment variables in cache hit rate computation. ([#422](https://github.com/rapidsai/cuvs/pull/422)) [@bdice](https://github.com/bdice)
+- Use dashes in cuvs-bench package name. ([#417](https://github.com/rapidsai/cuvs/pull/417)) [@bdice](https://github.com/bdice)
+- We need to enable the c_api by default ([#416](https://github.com/rapidsai/cuvs/pull/416)) [@robertmaynard](https://github.com/robertmaynard)
+- print sccache stats in builds ([#413](https://github.com/rapidsai/cuvs/pull/413)) [@jameslamb](https://github.com/jameslamb)
+- make conda installs in CI stricter ([#406](https://github.com/rapidsai/cuvs/pull/406)) [@jameslamb](https://github.com/jameslamb)
+- Ivf c example ([#404](https://github.com/rapidsai/cuvs/pull/404)) [@abner-ma](https://github.com/abner-ma)
+- Prune workflows based on changed files ([#392](https://github.com/rapidsai/cuvs/pull/392)) [@KyleFromNVIDIA](https://github.com/KyleFromNVIDIA)
+- [WIP] Add pinned memory resource to C API ([#311](https://github.com/rapidsai/cuvs/pull/311)) [@ajit283](https://github.com/ajit283)
+- Dynamic Batching ([#261](https://github.com/rapidsai/cuvs/pull/261)) [@achirkin](https://github.com/achirkin)
+
 # cuvs 24.10.00 (9 Oct 2024)
 
 ## 🐛 Bug Fixes

conda/environments/all_cuda-118_arch-aarch64.yaml

@@ -7,7 +7,7 @@ channels:
 - conda-forge
 - nvidia
 dependencies:
-- breathe
+- breathe>=4.35.0
 - c-compiler
 - clang
 - clang-tools=16.0.6
@@ -44,7 +44,6 @@ dependencies:
 - nvcc_linux-aarch64=11.8
 - openblas
 - pre-commit
-- pydata-sphinx-theme
 - pylibraft==25.2.*,>=0.0.0a0
 - pytest-cov
 - pytest==7.*
@@ -55,5 +54,8 @@ dependencies:
 - scikit-learn
 - sphinx-copybutton
 - sphinx-markdown-tables
+- sphinx>=8.0.0
 - sysroot_linux-aarch64==2.17
+- pip:
+  - nvidia-sphinx-theme
 name: all_cuda-118_arch-aarch64

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

@@ -7,7 +7,7 @@ channels:
 - conda-forge
 - nvidia
 dependencies:
-- breathe
+- breathe>=4.35.0
 - c-compiler
 - clang
 - clang-tools=16.0.6
@@ -44,7 +44,6 @@ dependencies:
 - nvcc_linux-64=11.8
 - openblas
 - pre-commit
-- pydata-sphinx-theme
 - pylibraft==25.2.*,>=0.0.0a0
 - pytest-cov
 - pytest==7.*
@@ -55,5 +54,8 @@ dependencies:
 - scikit-learn
 - sphinx-copybutton
 - sphinx-markdown-tables
+- sphinx>=8.0.0
 - sysroot_linux-64==2.17
+- pip:
+  - nvidia-sphinx-theme
 name: all_cuda-118_arch-x86_64

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

@@ -7,7 +7,7 @@ channels:
 - conda-forge
 - nvidia
 dependencies:
-- breathe
+- breathe>=4.35.0
 - c-compiler
 - clang
 - clang-tools=16.0.6
@@ -40,7 +40,6 @@ dependencies:
 - numpydoc
 - openblas
 - pre-commit
-- pydata-sphinx-theme
 - pylibraft==25.2.*,>=0.0.0a0
 - pytest-cov
 - pytest==7.*
@@ -51,5 +50,8 @@ dependencies:
 - scikit-learn
 - sphinx-copybutton
 - sphinx-markdown-tables
+- sphinx>=8.0.0
 - sysroot_linux-aarch64==2.17
+- pip:
+  - nvidia-sphinx-theme
 name: all_cuda-125_arch-aarch64

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

@@ -7,7 +7,7 @@ channels:
 - conda-forge
 - nvidia
 dependencies:
-- breathe
+- breathe>=4.35.0
 - c-compiler
 - clang
 - clang-tools=16.0.6
@@ -40,7 +40,6 @@ dependencies:
 - numpydoc
 - openblas
 - pre-commit
-- pydata-sphinx-theme
 - pylibraft==25.2.*,>=0.0.0a0
 - pytest-cov
 - pytest==7.*
@@ -51,5 +50,8 @@ dependencies:
 - scikit-learn
 - sphinx-copybutton
 - sphinx-markdown-tables
+- sphinx>=8.0.0
 - sysroot_linux-64==2.17
+- pip:
+  - nvidia-sphinx-theme
 name: all_cuda-125_arch-x86_64

conda/environments/bench_ann_cuda-118_arch-aarch64.yaml

@@ -18,6 +18,8 @@ dependencies:
 - cuda-python>=11.8.5,<12.0a0
 - cuda-version=11.8
 - cudatoolkit
+- cupy>=12.0.0
+- cuvs==25.2.*,>=0.0.0a0
 - cxx-compiler
 - cython>=3.0.0
 - dlpack>=0.8,<1.0
@@ -32,6 +34,7 @@ dependencies:
 - libcusolver=11.4.1.48
 - libcusparse-dev=11.7.5.86
 - libcusparse=11.7.5.86
+- libcuvs==25.2.*,>=0.0.0a0
 - librmm==25.2.*,>=0.0.0a0
 - matplotlib
 - nccl>=2.19

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

@@ -18,6 +18,8 @@ dependencies:
 - cuda-python>=11.8.5,<12.0a0
 - cuda-version=11.8
 - cudatoolkit
+- cupy>=12.0.0
+- cuvs==25.2.*,>=0.0.0a0
 - cxx-compiler
 - cython>=3.0.0
 - dlpack>=0.8,<1.0
@@ -32,6 +34,7 @@ dependencies:
 - libcusolver=11.4.1.48
 - libcusparse-dev=11.7.5.86
 - libcusparse=11.7.5.86
+- libcuvs==25.2.*,>=0.0.0a0
 - librmm==25.2.*,>=0.0.0a0
 - matplotlib
 - nccl>=2.19

conda/environments/bench_ann_cuda-125_arch-aarch64.yaml

@@ -19,6 +19,8 @@ dependencies:
 - cuda-profiler-api
 - cuda-python>=12.6.2,<13.0a0
 - cuda-version=12.5
+- cupy>=12.0.0
+- cuvs==25.2.*,>=0.0.0a0
 - cxx-compiler
 - cython>=3.0.0
 - dlpack>=0.8,<1.0
@@ -29,6 +31,7 @@ dependencies:
 - libcurand-dev
 - libcusolver-dev
 - libcusparse-dev
+- libcuvs==25.2.*,>=0.0.0a0
 - librmm==25.2.*,>=0.0.0a0
 - matplotlib
 - nccl>=2.19

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

@@ -19,6 +19,8 @@ dependencies:
 - cuda-profiler-api
 - cuda-python>=12.6.2,<13.0a0
 - cuda-version=12.5
+- cupy>=12.0.0
+- cuvs==25.2.*,>=0.0.0a0
 - cxx-compiler
 - cython>=3.0.0
 - dlpack>=0.8,<1.0
@@ -29,6 +31,7 @@ dependencies:
 - libcurand-dev
 - libcusolver-dev
 - libcusparse-dev
+- libcuvs==25.2.*,>=0.0.0a0
 - librmm==25.2.*,>=0.0.0a0
 - matplotlib
 - nccl>=2.19

conda/recipes/cuvs-bench-cpu/meta.yaml

@@ -59,6 +59,7 @@ requirements:
     - glog {{ glog_version }}
     - h5py {{ h5py_version }}
     - matplotlib
+    - numpy >=1.23,<3.0a0
     - pandas
     - pyyaml
     - python

conda/recipes/cuvs-bench/meta.yaml

@@ -79,6 +79,7 @@ requirements:
     - python
     - rapids-build-backend>=0.3.0,<0.4.0.dev0
     - rmm ={{ minor_version }}
+    - setuptools>=64.0.0
 
   run:
     - benchmark
@@ -88,10 +89,11 @@ requirements:
     - cudatoolkit
     {% else %}
     - cuda-cudart
+    - cupy>=12.0.0
     - libcublas
     {% endif %}
     - glog {{ glog_version }}
-    - libcuvs {{ version }}
+    - cuvs {{ version }}
     - h5py {{ h5py_version }}
     - matplotlib
     - pandas

cpp/CMakeLists.txt

@@ -486,13 +486,14 @@ if(BUILD_SHARED_LIBS)
                       "$<$<COMPILE_LANGUAGE:CUDA>:${CUVS_CUDA_FLAGS}>"
   )
   target_link_libraries(
-    cuvs_objs PUBLIC raft::raft rmm::rmm rmm::rmm_logger ${CUVS_CTK_MATH_DEPENDENCIES}
-                     $<TARGET_NAME_IF_EXISTS:OpenMP::OpenMP_CXX>
-              PRIVATE rmm::rmm_logger_impl
+    cuvs_objs
+    PUBLIC raft::raft rmm::rmm rmm::rmm_logger ${CUVS_CTK_MATH_DEPENDENCIES}
+           $<TARGET_NAME_IF_EXISTS:OpenMP::OpenMP_CXX>
+    PRIVATE rmm::rmm_logger_impl raft::raft_logger_impl
   )
 
   add_library(cuvs SHARED $<FILTER:$<TARGET_OBJECTS:cuvs_objs>,EXCLUDE,rmm.*logger>)
-  add_library(cuvs_static STATIC  $<FILTER:$<TARGET_OBJECTS:cuvs_objs>,EXCLUDE,rmm.*logger>)
+  add_library(cuvs_static STATIC $<FILTER:$<TARGET_OBJECTS:cuvs_objs>,EXCLUDE,rmm.*logger>)
 
   target_compile_options(
     cuvs INTERFACE $<$<COMPILE_LANG_AND_ID:CUDA,NVIDIA>:--expt-extended-lambda
@@ -704,7 +705,7 @@ target_compile_definitions(cuvs::cuvs INTERFACE $<$<BOOL:${CUVS_NVTX}>:NVTX_ENAB
     target_link_libraries(
       cuvs_c
       PUBLIC cuvs::cuvs ${CUVS_CTK_MATH_DEPENDENCIES}
-      PRIVATE raft::raft rmm::rmm_logger_impl
+      PRIVATE raft::raft rmm::rmm_logger_impl raft::raft_logger_impl
     )
 
     # ensure CUDA symbols aren't relocated to the middle of the debug build binaries

cpp/bench/ann/CMakeLists.txt

@@ -126,10 +126,11 @@ function(ConfigureAnnBench)
     PRIVATE ${ConfigureAnnBench_LINKS}
             nlohmann_json::nlohmann_json
             Threads::Threads
+            $<TARGET_NAME_IF_EXISTS:raft::raft_logger>
             $<$<BOOL:${GPU_BUILD}>:CUDA::cudart_static>
             $<TARGET_NAME_IF_EXISTS:OpenMP::OpenMP_CXX>
             $<TARGET_NAME_IF_EXISTS:conda_env>
-            $<TARGET_NAME_IF_EXISTS:cuvs_bench_rmm_logger>
+            $<TARGET_NAME_IF_EXISTS:cuvs_bench_logger>
   )
 
   set_target_properties(
@@ -175,9 +176,11 @@ function(ConfigureAnnBench)
   add_dependencies(CUVS_ANN_BENCH_ALL ${BENCH_NAME})
 endfunction()
 
-if(CUVS_FAISS_ENABLE_GPU)
-  add_library(cuvs_bench_rmm_logger OBJECT)
-  target_link_libraries(cuvs_bench_rmm_logger PRIVATE rmm::rmm_logger_impl)
+if(CUVS_FAISS_ENABLE_GPU OR CUVS_ANN_BENCH_SINGLE_EXE)
+  add_library(cuvs_bench_logger OBJECT)
+  target_link_libraries(
+    cuvs_bench_logger PRIVATE rmm::rmm_logger_impl $<TARGET_NAME_IF_EXISTS:raft::raft_logger>
+  )
 endif()
 
 # ##################################################################################################
@@ -303,8 +306,14 @@ if(CUVS_ANN_BENCH_SINGLE_EXE)
 
   target_link_libraries(
     ANN_BENCH
-    PRIVATE raft::raft nlohmann_json::nlohmann_json benchmark::benchmark dl fmt::fmt-header-only
-            spdlog::spdlog_header_only $<$<TARGET_EXISTS:CUDA::nvtx3>:CUDA::nvtx3> rmm::rmm_logger_impl
+    PRIVATE raft::raft
+            nlohmann_json::nlohmann_json
+            benchmark::benchmark
+            dl
+            fmt::fmt-header-only
+            spdlog::spdlog_header_only
+            $<$<TARGET_EXISTS:CUDA::nvtx3>:CUDA::nvtx3>
+            cuvs_bench_logger
   )
   set_target_properties(
     ANN_BENCH