CHANGELOG.md

rapids-cmake 26.06.00 (3 Jun 2026)

🚨 Breaking Changes

• Allow proprietary_binary-only packages without git_url/git_tag by @bdice in #1004
• Remove nvcomp package and proprietary binary infrastructure by @vyasr in #1013

🐛 Bug Fixes

• Fix CCCL_DIR cache corruption from CPM redirect stub by @bdice in #997
• export_cpm: Avoid saving CMake Package Redirection dir by @robertmaynard in #998
• Revert "export_cpm: Avoid saving CMake Package Redirection dir (#998)" by @KyleFromNVIDIA in #1000
• export_cpm: Better _DIR heuristic in the face of ambiguity by @robertmaynard in #1007
• Fix rapids_cpm_nvtx3 unconditional NVTX3_INSTALL and undefined exclude variable by @vyasr in #1009
• Update CCCL version by @KyleFromNVIDIA in #1015
• Update CCCL version by @bdice in #1018

📖 Documentation

• Update docs for proprietary_binary-only packages and nvCOMP by @bdice in #1006

🚀 New Features

• Update rapids-cmake to have a set of examples showing common patterns by @robertmaynard in #976
• Bump google benchmark version from 1.8.0 to 1.9.5 by @Jacobfaib in #992
• Add offline build example and docs by @robertmaynard in #995
• Upgrade to nvcomp 5.2.0.10 by @bdice in #1003
• Support EXCLUDE_FROM_ALL in all rapids_cpm_* functions by @vyasr in #1010

🛠️ Improvements

• chore: bump codespell version for Python 3.14+ compatibility by @gforsyth in #989
• Update CCCL to 3.4.0 pre-release by @bdice in #990
• Update CCCL to 3.4.0 pre-release by @bdice in #996
• Update to clang 20.1.8 by @bdice in #1001
• Update CCCL version by @nirandaperera in #1008
• Support tarball-mode packages in patch application by @bdice in #1011
• fix(ci): resolve all zizmor findings and add zizmor pre-commit checks by @gforsyth in #1016
• Build and test with CUDA 13.2.0 by @bdice in #1019
• Use GitHub tarball downloads for CPM dependencies by @bdice in #1005
• fix(ci): add explicit actions: write permission for telemetry-summarize by @gforsyth in #1020
• Update CCCL version by @bdice in #1022

Full Changelog: https://github.com/rapidsai/rapids-cmake/compare/v26.06.00a...release/26.06

rapids-cmake 26.04.00 (8 Apr 2026)

🚨 Breaking Changes

• Update CCCL to 3.3.0 pre-release by @bdice in #981

🐛 Bug Fixes

• FINAL_CODEBLOCK can see targets that should be global by @robertmaynard in #993

🚀 New Features

• Add URL option to fetch packages from GitHub tarballs by @bdice in #968
• CMake 3.30+ means CMAKE_PROJECT_INCLUDE can be a list by @robertmaynard in #984
• Update cuco version to fetch the cuda fancy iterators by @PointKernel in #987

🛠️ Improvements

• Forward-merge release/26.02 into main by @bdice in #974
• Drop Python 3.10 support by @gforsyth in #975
• Use verify-hardcoded-version pre-commit hook by @KyleFromNVIDIA in #977
• Don't hard-code branch names by @KyleFromNVIDIA in #978
• Update CCCL to 3.2.1 pre-release by @bdice in #979
• Add support for Python 3.14 by @gforsyth in #982

Full Changelog: https://github.com/rapidsai/rapids-cmake/compare/v26.04.00a...release/26.04

rapids-cmake 26.02.00 (4 Feb 2026)

🚨 Breaking Changes

• Default to static linking of libcudart by @bdice in #949
• Update to CCCL v3.2.0 by @bdice in #955

🐛 Bug Fixes

• Update to a newer CCCL 3.2.x prerelease with a fix for environment queries by @bdice in #967

🚀 New Features

• Add support for finding CMake prefixes exported by Python packages by @vyasr in #951

🛠️ Improvements

• Use strict priority in CI conda tests by @bdice in #942
• Use strict priority in CI conda tests by @bdice in #947
• Upgrade to nvCOMP 5.1.0.21 by @bdice in #946
• Enable merge barriers by @KyleFromNVIDIA in #948
• feat(SABI): use stable abi naming when requested by @gforsyth in #954
• Bump cuco version to fetch HLL fix by @PointKernel in #952
• Clarify docs for rapids_cuda_init_architectures() by @KyleFromNVIDIA in #956
• Add CUDA 13.1 support by @bdice in #953
• Bump cuco version to fetch new HLL add_if API and bug fixes by @PointKernel in #958
• build and test against CUDA 13.1.0 by @jameslamb in #960
• Update to a newer CCCL 3.2.x prerelease with a fix for mdspan by @bdice in #959
• Update to a newer CCCL 3.2.x prelease by @nirandaperera in #961
• Use main shared-workflows branch by @jameslamb in #962
• Cherry-pick 25.12 changelog to main by @AyodeAwe in #963
• Fix update-version.sh to not modify cmake_minimum_required by @AyodeAwe in #964
• fix(build): build package on merge to release/* branch by @gforsyth in #971
• Update CCCL to final v3.2.0 release by @bdice in #970

Full Changelog: https://github.com/rapidsai/rapids-cmake/compare/v26.02.00a...release/26.02

rapids-cmake 25.12.00 (10 Dec 2025)

🚨 Breaking Changes

• Remove deprecated support for fmt and spdlog by @bdice in #914
• Remove deprecated support for CUDA 11 by @bdice in #913
• Remove spdlog and fmt and move to rapids-logger 0.2.1 by @vyasr in #918
• Require CUDA 12.2+ by @jakirkham in #934

🐛 Bug Fixes

• Fix RAPIDS_BRANCH file by @KyleFromNVIDIA in #910
• Temporarily restore spdlog and fmt by @vyasr in #917
• Update rapids-logger to 0.2.2 by @vyasr in #920
• Can't use Genex in GLOBAL_TARGETS by @robertmaynard in #921
• Update rapids-logger to 0.2.3 by @vyasr in #923
• refactored update-version.sh to handle new branching strategy by @rockhowse in #936

🚀 New Features

• Adding ENABLE_UNSTABLE option for rapids_cpm_cccl by @nirandaperera in #919

🛠️ Improvements

• Update nvbench by @bdice in #895
• Update to CCCL v3.1.0 by @bdice in #906
• Upgrade to CPM 0.42.0 by @vyasr in #916
• Forward-merge branch-25.10 into branch-25.12 by @bdice in #927
• Update CCCL to final v3.1.0 release tag by @bdice in #928
• Use main in RAPIDS_BRANCH by @bdice in #930
• Use main branch by @bdice in #931
• Use main shared-workflows branch by @bdice in #932
• Use SPDX for all copyright headers by @KyleFromNVIDIA in #933
• Bump cuco version to fetch stream-ordered allocator by @PointKernel in #929
• Bump cuco version to bypass equal check for multiset/multimap by @PointKernel in #935
• Some more SPDX fixes by @KyleFromNVIDIA in #937
• Update CCCL to get backports for XGBoost compatibility by @bdice in #944

New Contributors

• @nirandaperera made their first contribution in #919
• @rockhowse made their first contribution in #936

Full Changelog: https://github.com/rapidsai/rapids-cmake/compare/v25.12.00a...release/25.12

rapids-cmake 25.10.00 (8 Oct 2025)

🚨 Breaking Changes

• Upgrade to nvCOMP 5.0.0.6 (#896) @vuule

🐛 Bug Fixes

• Fix generation of RAPIDS_BRANCH (#909) @KyleFromNVIDIA
• change Architectures to avoid compute_100/120 code gen bug (#900) @robertmaynard
• set_architectures + NATIVE works even without a GPU (#899) @robertmaynard
• set_architectures correctly limits CUDA 12.8 to sm_120 (#898) @robertmaynard
• rapids_export handles PROJECT_VERSION CMake 4.1+ change in behavior (#893) @robertmaynard
• Fix detect architectures to support gcc 14 (#892) @ChuckHastings
• Update RAPIDS.cmake to match new logic in other projects (#884) @robertmaynard
• Update RAPIDS_BRANCH to 25.10 (#883) @robertmaynard
• Expand versions.json to have src_dir representation (#882) @robertmaynard

🚀 New Features

• Update cuco version to fetch signed comparison warning fix (#905) @PointKernel
• Updated dependencies.yaml to support CUDA 13 (#901) @robertmaynard
• Use family arch flags whenever possible (#897) @robertmaynard
• Upgrade to nvCOMP 5.0.0.6 (#896) @vuule
• Update set_architectures to support CUDA 13 (#891) @robertmaynard
• rapids-cmake Expose rapids_cuda_enable_fatbin_compression to have better CUDA 12/13 compression API (#890) @robertmaynard
• CUDA 13 builds will properly pull nvcomp 4.2.0.11 (#889) @robertmaynard
• Properly store source_subdir values from rapids_cpm_find calls (#885) @robertmaynard

🛠️ Improvements

• Configure repo for automatic release notes generation (#907) @AyodeAwe
• Use branch-25.10 again (#904) @jameslamb
• Update rapids-dependency-file-generator (#903) @KyleFromNVIDIA
• Update cuco version to fetch the new roaring bitmap (#902) @PointKernel
• Bump cuco version to fetch a bug fix for device retrieve APIs (#887) @PointKernel
• chore: update docker version tags (#873) @gforsyth

rapids-cmake 25.08.00 (6 Aug 2025)

🚨 Breaking Changes

• rapids_cpm_cccl: Remove support for CCCL < 2.8 (#859) @robertmaynard
• Remove CUDA 11 support (#855) @KyleFromNVIDIA
• Update to CCCL 3.0 (#854) @vyasr
• Require cpp subdirectory for RMM (#832) @bdice

🐛 Bug Fixes

• CCCL: disable PDL (#876) @bdice
• Use RMM main (new branching strategy) to fix downstream fetching issues (#862) @bdice
• rapids_cpm_cccl: Update to new location of cccl-config (#858) @robertmaynard
• Remove CCCL patches that aren't used anymore (#857) @robertmaynard
• Fetch the atomic fix in CCCL 3.0 (#856) @PointKernel

📖 Documentation

• add docs on CI workflow inputs (#868) @jameslamb

🚀 New Features

• Update CCCL version tag for PDL disable (#879) @davidwendt
• Use RAPIDS_BRANCH file to handle the new branching strategy (#870) @robertmaynard
• rapids-cmake: Add support for a version suffix to mean using main (#864) @robertmaynard
• rapids_cpm_cccl: Remove support for CCCL < 2.8 (#859) @robertmaynard
• Update to CCCL 3.0 (#831) @bdice

🛠️ Improvements

• Update to CCCL v3.0.2 (#878) @bdice
• fix(docker): use versioned -latest tag for all rapidsai images (#871) @gforsyth
• Revert "Use RMM main (new branching strategy)" (#869) @robertmaynard
• Rename *.hpp.in to *.h.in to signify that they are C headers (#867) @KyleFromNVIDIA
• refactor(shellcheck): enable for all files (#866) @gforsyth
• Remove nvidia and dask channels (#865) @vyasr
• Upgrade cuCollections to fetch the new storage for better runtime performance (#861) @PointKernel
• Remove CUDA 11 support (#855) @KyleFromNVIDIA
• Update to CCCL 3.0 (#854) @vyasr
• Deprecate fmt and spdlog (#853) @vyasr
• Forward-merge branch-25.06 into branch-25.08 (#848) @gforsyth
• Update to NVTX 3.2.0. (#844) @bdice
• Forward-merge branch-25.06 into branch-25.08 (#839) @gforsyth
• Temporarily use patched CCCL (#833) @bdice
• Require cpp subdirectory for RMM (#832) @bdice

rapids-cmake 25.06.00 (5 Jun 2025)

🚨 Breaking Changes

• Make RAPIDS.cmake inlineable without requiring updates (#825) @KyleFromNVIDIA
• Make RAPIDS.cmake inlineable without requiring updates (#822) @KyleFromNVIDIA
• Add build patch functionality (#801) @KyleFromNVIDIA

🐛 Bug Fixes

• Get newer CCCL 2.8 with fix for atomics. (#840) @bdice
• Unset CUDAARCHS in test (#835) @KyleFromNVIDIA
• Revert "Make RAPIDS.cmake inlineable without requiring updates" (#824) @bdice
• Add include_guard(GLOBAL) to download_with_retry.cmake (#818) @KyleFromNVIDIA
• Fix issue with how project pins are determined (#815) @KyleFromNVIDIA
• Disable git_shallow for nvtx3 and bs_thread_pool (#811) @KyleFromNVIDIA
• Fix NVTX3 export (#804) @KyleFromNVIDIA

🚀 New Features

• Upgrade nvbench for latest throttling control updates (#830) @PointKernel
• Add build patch functionality (#801) @KyleFromNVIDIA

🛠️ Improvements

• explicitly provide 'script' to CI workflows (#849) @jameslamb
• Finish CUDA 12.9 migration and use branch-25.06 workflows (#847) @bdice
• Update to clang 20 (#846) @bdice
• Update to CCCL 2.8.4. (#845) @bdice
• Bump cuco version to fetch multiple CCCL upgrade fixes (#843) @PointKernel
• Build and test with CUDA 12.9.0 (#842) @bdice
• Upgrade cuco version to include murmurhash performance fix (#838) @PointKernel
• feat: add Python 3.13 support (#834) @gforsyth
• Update cuCollections to avoid CCCL deprecation warnings (#828) @miscco
• Use cpp subdirectory for RMM. (#827) @bdice
• Make RAPIDS.cmake inlineable without requiring updates (#825) @KyleFromNVIDIA
• Remove hard-coded RAPIDS versions (#823) @KyleFromNVIDIA
• Make RAPIDS.cmake inlineable without requiring updates (#822) @KyleFromNVIDIA
• Format all CMake code (#821) @bdice
• In conda-build produced binaries, use paths relative to $PREFIX (#817) @jakirkham
• Remove CMP0135 policy checks. (#816) @bdice
• Use hashes instead of tags (#812) @KyleFromNVIDIA
• refactor(conda): remove unused conda package and associated CI resources (#810) @gforsyth
• Allow retrying downloads on network failures (#809) @pentschev
• Define -DCUB_DISABLE_NAMESPACE_MAGIC and -DCUB_IGNORE_NAMESPACE_MAGIC_ERROR in rapids_cpm_cccl (#807) @trxcllnt
• Update to CCCL 2.8.3 (#793) @bdice

rapids-cmake 25.04.00 (9 Apr 2025)

🚨 Breaking Changes

• Backport build patch functionality (#805) @KyleFromNVIDIA
• Update gtest version (#795) @vyasr
• Remove now superfluous fmt patch (#790) @vyasr
• Use CTest dynamic resource spec generation (#786) @KyleFromNVIDIA
• Switch to using rapids-logger as a library (#765) @vyasr
• Require CMake 3.30.4, and latest CPM (#753) @robertmaynard

🐛 Bug Fixes

• Set correct compiler feature for generate_ctest_json (#802) @KyleFromNVIDIA
• Also download fmt when building nvbench statically (#792) @vyasr
• Rollback to CPM 0.40.0 (#789) @robertmaynard
• Sanitize patch inputs so that quotes are properly escaped (#787) @robertmaynard
• Fix tests broken by rapids-logger changes (#775) @vyasr

🚀 New Features

• Backport build patch functionality (#805) @KyleFromNVIDIA
• Add support for components in Cython rapids_cython_create_modules (#783) @pentschev
• rapids_cpm_package_override: Always load overrides (#778) @robertmaynard
• Require CMake 3.30.4, and latest CPM (#753) @robertmaynard

🛠️ Improvements

• Update gtest version (#795) @vyasr
• Remove now superfluous fmt patch (#790) @vyasr
• Use CTest dynamic resource spec generation (#786) @KyleFromNVIDIA
• Use conda-build instead of conda-mambabuild (#784) @bdice
• Set -Wno-deprecated-gpu-targets when compiling for RAPIDS architectures (#776) @bdice
• Add shellcheck to pre-commit and fix warnings (#772) @gforsyth
• Update cuCollection to not rely on thrust::identity (#771) @miscco
• Use rapids-generate-version for package versions (#770) @bdice
• Use shared-workflows branch-25.04 (#768) @bdice
• add telemetry (#767) @msarahan
• Add build_type input field for test.yaml (#766) @gforsyth
• Switch to using rapids-logger as a library (#765) @vyasr
• Add verify-codeowners hook (#762) @KyleFromNVIDIA
• Forward-merge branch-25.02 to branch-25.04 (#760) @bdice
• Migrate to NVKS for amd64 CI runners (#758) @bdice

rapids-cmake 25.02.00 (13 Feb 2025)

🚨 Breaking Changes

• Update to CCCL 2.7.0-rc2. (#710) @bdice

🐛 Bug Fixes

• Sanitize patch input to protect against CMake Language limitations (#773) @robertmaynard
• Backport cccl PR 3578 to v2.7 to unblock CUDA 12.8 builds (#756) @robertmaynard
• Use sysroot 2.17 on CUDA < 11.8. (#745) @bdice
• rapids_cpm_cccl use proper CCCL version value to compute install rules (#742) @robertmaynard
• Correct nvbench + conda test to explicitly pull spdlog (#733) @robertmaynard

📖 Documentation

• fix signature in rapids_export() docs (#732) @jameslamb

🚀 New Features

• Add support for new architectures in CUDA 12.8 (#746) @robertmaynard
• Update logger hash (#738) @vyasr
• Add rapids-logger (#737) @vyasr
• Fea/support embedded patch content (#728) @robertmaynard
• Update to CCCL 2.7.0-rc2. (#710) @bdice

🛠️ Improvements

• Revert CUDA 12.8 shared workflow branch changes (#759) @vyasr
• Build and test with CUDA 12.8.0 (#754) @bdice
• Allow sysroot 2.28 in CUDA 11 jobs. (#749) @bdice
• Fetch cuco filter alignment fix to unblock RAPIDS CI (#744) @PointKernel
• Use CCCL v2.7.0 final tag. (#743) @bdice
• Use GCC 13 in CUDA 12 conda builds. (#741) @bdice
• Add new global targets for rmm (#739) @vyasr
• remove setup.cfg, use pyproject.toml (#736) @jameslamb
• Bump cuco by one commit to include cuco::arrow_filter_policy updates (#735) @mhaseeb123
• Require Cython to be found (#734) @vyasr
• Update version references in workflow (#729) @AyodeAwe
• Bump cuco version to fetch various fixes and new features (#723) @PointKernel
• Add breaking change workflow trigger (#719) @AyodeAwe

rapids-cmake 24.12.00 (11 Dec 2024)

🐛 Bug Fixes

• Fix failures in cpm_generate_pins-nested tests (#724) @vyasr
• Search for the normalized package name when patching (#715) @vyasr
• rapids_export_write_language no longer generates cmake dev warnings (#712) @robertmaynard
• Update rapids-cmake to support cccl 2.8 new install rules (#711) @robertmaynard
• gtest_discover_tests doesn't crash rapids_test_install_relocatable (#697) @robertmaynard

📖 Documentation

• Fix docs warning about get_html_theme_path. (#705) @bdice

🚀 New Features

• Verifies that current_json_dir placeholder works with normalized package names (#716) @robertmaynard
• Allow for cpm override via file and direct calls to rapids_cpm_package_override (#714) @robertmaynard
• Upgrade nvcomp to 4.1.0.6 (#709) @bdice
• Warn when override of a default package uses the incorrect case (#708) @robertmaynard
• Allow CCCL 2.7 to be used in a rapids-cmake override (#706) @robertmaynard
• version.json patch command supports required field (#700) @robertmaynard
• Bump cuco version to fetch the Hyperloglog renaming (#698) @PointKernel

🛠️ Improvements

• Bump cuco version to fetch the new multiset retrieve (#707) @PointKernel

rapids-cmake 24.10.00 (9 Oct 2024)

🚨 Breaking Changes

• update fmt (to 11.0.2) and spdlog (to 1.14.1) (#689) @jameslamb
• Remove 24.10 deprecated commands (#665) @robertmaynard

🐛 Bug Fixes

• Add workaround for gcc bug in cuco (#685) @miscco
• Revert "Update to CPM 0.40.2 to fix CMake 3.30 deprecation warnings (#678)" (#679" (#679)) @jameslamb
• Update to CPM 0.40.2 to fix CMake 3.30 deprecation warnings (#678) @robertmaynard
• Bump cuco version to fetch several bug fixes (#677) @PointKernel
• rapids_cpm_nvcomp has consistent behavior on every cmake execution (#676) @robertmaynard
• rapids-cmake generated C++ files have current copyright year (#674) @robertmaynard
• write_language now properly walks up the add_subdirectory call stack (#671) @robertmaynard

📖 Documentation

• Update docs for overriding rapids-cmake (#681) @KyleFromNVIDIA
• add ASSOCIATED_TARGETS to function signature in rapids_cython_create_modules() docs (#670) @jameslamb

🚀 New Features

• Reduce cpm tests network usage (#683) @robertmaynard
• rapids-cmake tests no longer download CPM when NO_CPM_CACHE is set. (#682) @robertmaynard
• Remove deprecated rapids_export_find_package_* signatures (#666) @robertmaynard
• Upgrade nvcomp to 4.0.1 (#633) @vuule

🛠️ Improvements

• Add INSTALL_TARGET argument to rapids_add_test() (#692) @KyleFromNVIDIA
• Use CI workflow branch 'branch-24.10' again (#691) @jameslamb
• update fmt (to 11.0.2) and spdlog (to 1.14.1) (#689) @jameslamb
• Add support for Python 3.12 (#688) @jameslamb
• Update rapidsai/pre-commit-hooks (#686) @KyleFromNVIDIA
• Drop Python 3.9 support (#684) @jameslamb
• Allow nested lib location for nvcomp (#675) @KyleFromNVIDIA
• Update pre-commit hooks (#669) @KyleFromNVIDIA
• Improve update-version.sh (#668) @bdice
• Remove 24.10 deprecated commands (#665) @robertmaynard

rapids-cmake 24.08.00 (7 Aug 2024)

🚨 Breaking Changes

• Move required CMake version to 3.26.4 (#627) @robertmaynard
• Removes legacy rapids-cmake cython implementations as it is deprecated in 24.08 (#614) @robertmaynard
• Update CCCL to v2.5.0 (#607) @trxcllnt

🐛 Bug Fixes

• bs_thread_pool uses C++ 17 (#662) @KyleFromNVIDIA
• Use CMAKE_CUDA_ARCHITECTURES value when ENV{CUDAARCHS} is set (#659) @robertmaynard
• Pass GLOBAL_TARGETS to rapids_cpm_find() in rapids_cpm_bs_thread_pool() (#655) @KyleFromNVIDIA
• Add rapids_cpm_nvtx3 to cmake-format-rapids-cmake.json (#652) @KyleFromNVIDIA
• generate_resource_spec uses the enabled languages to determine compiler (#645) @robertmaynard
• Set CUDA_RUNTIME_LIBRARY to documented case style (#641) @KyleFromNVIDIA
• rapids_test_install_relocatable now handles SOVERSION target properties (#636) @robertmaynard
• Eval CMAKE_CUDA_ARCHITECTURES before ENV{CUDAARCHS} (#624) @robertmaynard

📖 Documentation

• Fix CPM package docs. (#637) @bdice
• expand rapids-cmake cpm docs (#613) @robertmaynard

🚀 New Features

• Add rapids_cpm_bs_thread_pool() (#651) @KyleFromNVIDIA
• proprietary_binary_cuda_version_mapping allows for better CUDA version mapping (#648) @robertmaynard
• use latest cuco with insert_or_apply CAS fix (#646) @srinivasyadav18
• Fetch the latest cuco with CAS fixes (#644) @PointKernel
• Update to CPM v0.40.0 (#642) @robertmaynard
• Remove CCCL patch for PR 211. (#640) @bdice
• Try any failing tests up to 3 times to guard against network issues (#639) @robertmaynard
• rapids_cmake_support_conda_env adds -O0 to debug compile lines (#635) @robertmaynard
• Update cuco git tag to fetch new multiset data structure (#628) @PointKernel
• Move required CMake version to 3.26.4 (#627) @robertmaynard
• Removes legacy rapids-cmake cython implementations as it is deprecated in 24.08 (#614) @robertmaynard
• Update CCCL to v2.5.0 (#607) @trxcllnt

🛠️ Improvements

• Use workflow branch 24.08 again (#647) @KyleFromNVIDIA
• Build and test with CUDA 12.5.1 (#643) @KyleFromNVIDIA
• Bump CCCL version to include cuda::std::span fix (#631) @sleeepyjack
• Adopt CI/packaging codeowners (#629) @bdice

rapids-cmake 24.06.00 (5 Jun 2024)

🐛 Bug Fixes

• Only output CUDA architectures we are building for once (#621) @robertmaynard
• set_architectures output correct CUDA arch values for RAPIDS mode (#619) @robertmaynard
• Always offer the install target names for nvtx3 (#617) @robertmaynard
• Support CMAKE_INSTALL_MESSAGE + rapids_test_install_relocatable (#604) @robertmaynard
• Ensure nvcomps build and install layouts are consistent (#602) @robertmaynard
• Correctly set the install location for nvcomp when using the proprietary binary (#597) @vyasr
• Ensure support_conda_env uses isystem (#588) @robertmaynard
• Update rapids_test_install_relocatable to be aware of CMake 3.29 (#586) @robertmaynard
• rapids_cpm_gtest(BUILD_STATIC) now doesn't search for a local version (#585) @robertmaynard
• Add new patch to hide more CCCL APIs (#580) @vyasr
• Forward-merge branch-24.04 into branch-24.06 [skip ci] (#565) @rapids-bot[bot]

📖 Documentation

• use inline code formatting in docs for variables, functions, and modules (#591) @jameslamb
• clarify language around how FetchContent is used in RAPIDS.cmake (#590) @jameslamb
• fix docs for rapids_export_ functions (#589) @jameslamb
• Better explain OVERRIDE option (#578) @robertmaynard

🚀 New Features

• Output what cuda archs rapids-cmake is building for (#609) @robertmaynard
• Add rapids_cpm_nvtx3. (#606) @bdice
• Refactor the common verify cpm pin test logic to a single source (#601) @robertmaynard
• rapids-cmake allow GENERATE_PINNED_VERSIONS via CMake variable (#600) @robertmaynard
• Allow for cpm default and override files via variables (#596) @robertmaynard
• Expand rapids_cpm_init to support custom default version files (#595) @robertmaynard
• Bump NVBench version for new main hooks. (#584) @alliepiper
• rapids_cython_create_modules(): Generate Cython dependency file (#579) @Jacobfaib
• rapids_cpm_gtest adds support for BUILD_STATIC (#576) @robertmaynard

🛠️ Improvements

• Fix nvtx3 build export (#615) @trxcllnt
• limit pinning tests to CPM-downloaded projects (#599) @jameslamb
• Migrate to {{ stdlib("c") }} (#594) @hcho3
• resolve 'file_key' deprecation warning from rapids-dependency-file-generator (#593) @jameslamb
• Prevent path conflict in builds (#571) @AyodeAwe
• Bump cuco version to fetch the latest set retrieve API (#569) @PointKernel
• Forward-merge branch-24.04 to branch-24.06 (#563) @bdice

rapids-cmake 24.04.00 (10 Apr 2024)

🐛 Bug Fixes

• nvcomp try proprietary binary when 'always_download' is on (#570) @robertmaynard
• Update pre-commit-hooks to v0.0.3 (#566) @KyleFromNVIDIA
• Always link against cudart_static (#564) @KyleFromNVIDIA
• Use CUDA compiler if available for generate_resource_spec (#561) @KyleFromNVIDIA
• rapids-cmake support empty patches array in json (#559) @robertmaynard
• Handle CMake 3.28 new EXCLUDE_FROM_ALL option of FetchContent (#557) @robertmaynard
• Add rapids_cuda_set_runtime to default includes of cuda (#538) @robertmaynard
• add rapids-dependency-file-generator pre-commit hook (#531) @jameslamb

🚀 New Features

• Deprecate rapids_cpm_libcudacxx and rapids_cpm_thrust. (#560) @bdice
• rapids_cpm_package_details now validates required entries exist (#558) @robertmaynard
• Support getting rapids-cmake via git clone (#555) @robertmaynard
• Bump nvbench version for faster benchmark runs (#549) @PointKernel
• Remove unneeded whitespace from json (#544) @robertmaynard
• Officially support env var expansion in version.json (#540) @robertmaynard
• rapids-cmake can generate pinned versions file (#530) @robertmaynard
• Fetch the latest cuco and remove outdated patches (#526) @PointKernel
• Support CUDA 12.2 (#521) @jameslamb

🛠️ Improvements

• Use conda env create --yes instead of --force (#573) @bdice
• Replace local copyright check with pre-commit-hooks verify-copyright (#556) @KyleFromNVIDIA
• Add patch to fix fmt v10.1.1 version (#548) @trxcllnt
• Add support for Python 3.11 (#547) @jameslamb
• Forward-merge branch-24.02 to branch-24.04 (#545) @bdice
• target branch-24.04 for GitHub Actions workflows (#541) @jameslamb
• Build generate_ctest_json in try_compile() (#537) @KyleFromNVIDIA
• Ensure that ctest is called with --no-tests=error. (#535) @bdice
• Update ops-bot.yaml (#532) @AyodeAwe

rapids-cmake 24.02.00 (12 Feb 2024)

🚨 Breaking Changes

• Drop Pascal architecture (60). (#482) @bdice

🐛 Bug Fixes

• Error out if generate_ctest_json fails to build or run (#533) @KyleFromNVIDIA
• rapids_cpm_cccl now works as expected when given DOWNLOAD_ONLY ON (#527) @robertmaynard
• Always download repos when they are being patched (#525) @vyasr
• Mark all cccl and cuco kernels with hidden visibility (#523) @robertmaynard
• Fix message context (#520) @vyasr
• Generate template copyright year at build time. (#325) (#519 (#519)) @KyleFromNVIDIA
• Add link libraries to generate_resource_spec cmake (#516) @davidwendt
• rapids_cpm_cccl preserve install location details from first invocation. (#513) @robertmaynard
• Only apply install rules for CCCL if we actually downloaded (#507) @bdice
• Mark flaky test as serial (#506) @vyasr
• Manually invoke install rules for components (#505) @vyasr
• multiple entry overrides now sets FetchContent for all entries (#494) @robertmaynard
• Remove deprecated function usages (#484) @robertmaynard

📖 Documentation

• Fix docs references to API sections. (#509) @bdice
• fix typo in README (#501) @jameslamb
• Fix indentation typo (#497) @vyasr

🚀 New Features

• rapids cpm patches now support differences in white space. (#515) @robertmaynard
• Upgrade nvCOMP to 3.0.5 (#498) @davidwendt
• Move to latest nvbench which has nvml+static support (#488) @robertmaynard
• Update to spdlog 1.12 and fmt 10.1.1 (#473) @kkraus14
• Support scikit-build-core (#433) @vyasr

🛠️ Improvements

• Remove usages of rapids-env-update (#524) @KyleFromNVIDIA
• refactor CUDA versions in dependencies.yaml (#517) @jameslamb
• Remove scikit-build from dependency list (#512) @vyasr
• Add patch reverting CCCL PR 211. (#511) @bdice
• Update cuCollections for CCCL 2.2.0 support. (#510) @bdice
• Disable NVBench CUPTI support by default. (#504) @bdice
• Remove CCCL::Thrust from GLOBAL_TARGETS. (#500) @bdice
• Add missing nvcomp targets (#496) @vyasr
• Add rapids_cpm_cccl feature. (#495) @bdice
• Drop Pascal architecture (60). (#482) @bdice

rapids-cmake 23.12.00 (6 Dec 2023)

🚨 Breaking Changes

• Upgrade nvCOMP to 3.0.4 (#451) @vuule

🐛 Bug Fixes

• Ensure nvbench initializes nvml context when built statically (#486) @robertmaynard
• Remove invalid argument to find_package_root (#483) @robertmaynard
• target from write_git_revision now works with export sets (#474) @robertmaynard

🚀 New Features

• Support static builds of gbench and nvbench. (#481) @robertmaynard
• Allow rapids_test to be used without CUDAToolkit (#480) @robertmaynard
• Update cuco git tag (#479) @sleeepyjack
• GTest will always be PIC enabled when built (#477) @robertmaynard
• Thrust when exported now automatically calls thrust_create_target (#467) @robertmaynard
• Upgrade nvCOMP to 3.0.4 (#451) @vuule

🛠️ Improvements

• Build concurrency for nightly and merge triggers (#490) @bdice
• Add patch for libcudacxx memory resource. (#476) @bdice
• Use branch-23.12 workflows. (#472) @bdice
• Express Python version in dependencies.yaml. (#470) @bdice
• Build CUDA 12.0 ARM conda packages. (#468) @bdice
• Update libcudacxx to 2.1.0 (#464) @bdice

rapids-cmake 23.10.00 (11 Oct 2023)

🐛 Bug Fixes

• Quote the list of patch files in case they have spaces in their paths (#463) @ericniebler
• cpm overrides don't occur when CPM_<pkg>_SOURCE exists (#458) @robertmaynard
• Use conda mambabuild not mamba mambabuild (#457) @bdice
• Support fmt use in debug builds (#456) @robertmaynard

📖 Documentation

• Move rapids_cpm_package_override to CPM section of docs (#462) @robertmaynard
• Improve docs around fetch content and rapids-cmake overrides (#444) @robertmaynard

🚀 New Features

• Bump cuco version (#452) @PointKernel

🛠️ Improvements

• Update image names (#461) @AyodeAwe
• Update to CPM v0.38.5 (#460) @trxcllnt
• Update to clang 16.0.6. (#459) @bdice
• Use copy-pr-bot (#455) @ajschmidt8

rapids-cmake 23.08.00 (9 Aug 2023)

🐛 Bug Fixes

• Use < gcc-11 with cuda 11.5 to avoid nvbench compile failures (#448) @robertmaynard
• Ensure tests the modify same git repo don't execute at the same time (#446) @robertmaynard
• Fix CUDA 11.5 tests by adding dependencies entries. (#443) @bdice
• Remove trailing comma and add pre-commit hook for JSON validation. (#440) @bdice
• When nvcomp is found locally print where it is on disk (#434) @robertmaynard
• Correct two issues found when testing CMake 3.27 rc2 (#432) @robertmaynard
• Correct re-root controls from conda-forge with thrust/cub/etc (#431) @robertmaynard
• Bug/proprietary binary obeys always_download (#430) @robertmaynard
• Correct install_relocatable issues found by libcudf (#423) @robertmaynard
• test_install_relocatable correct run_gpu_test.cmake location (#420) @robertmaynard
• Fea/move to latest nvbench (#417) @robertmaynard
• Use [@loader_path instead of $ORIGIN on MacOS (#403) @manopapad](https://github.com/loader_path instead of $ORIGIN on MacOS (#403) @manopapad)
• Make NAMESPACE property truly optional in rapids_export (#358) @agirault

🚀 New Features

• Update rapids-cmake ci to support conda-forge CUDA 12 (#437) @robertmaynard
• Bump cuco version (#435) @PointKernel
• Add rapids_cuda_set_runtime (#429) @robertmaynard
• support_conda_env support host and build CTK 12 locations (#428) @robertmaynard
• rapids_find_generate_module Support user code blocks (#415) @robertmaynard
• Rewrite of rapids_test_install_relocatable to support genex expressions (#410) @robertmaynard

🛠️ Improvements

• Conditionally modify envvar vs. global CMAKE_PREFIX_PATH in rapids_cmake_support_conda_env (#439) @trxcllnt
• Migrate to updated shared-action-workflows name for CUDA 12 CI (#438) @bdice
• Fix google benchmark name and update version (#425) @vyasr
• use rapids-upload-docs script (#419) @AyodeAwe
• Remove documentation build scripts for Jenkins (#418) @ajschmidt8
• Upload conda packages for rapids_core_dependencies. (#414) @bdice

rapids-cmake 23.06.00 (7 Jun 2023)

🚨 Breaking Changes

• Using deprecated CUDA_ARCHITECTURE values now produces an error. (#397) @robertmaynard
• rapids_cpm cccl packages cmake files are now relocated to not clash with upstream (#393) @robertmaynard

🐛 Bug Fixes

• Revert "Define Cython language_level explicitly. (#394)" (#396" (#396)) @vyasr
• rapids_cpm cccl packages cmake files are now relocated to not clash with upstream (#393) @robertmaynard

📖 Documentation

• Correct basics to api cross refs (#405) @robertmaynard

🚀 New Features

• Update cuco git tag to support cuco::static_set (#407) @PointKernel
• Upgrade GTest version to 1.13 (#401) @robertmaynard
• Using deprecated CUDA_ARCHITECTURE values now produces an error. (#397) @robertmaynard

🛠️ Improvements

• run docs nightly too (#413) @AyodeAwe
• Update cuco git tag to fetch several bug fixes (#412) @PointKernel
• Revert shared workflows branch (#406) @ajschmidt8
• Upgrade to Python 3.9 (drop Python 3.9) (#404) @shwina
• Remove usage of rapids-get-rapids-version-from-git (#402) @jjacobelli
• Update clang-format (#398) @bdice
• Define Cython language_level explicitly. (#394) @bdice

rapids-cmake 23.04.00 (6 Apr 2023)

🐛 Bug Fixes

• install_relocatable only installs files that exist (#392) @robertmaynard
• Revert "install tests environment properties (#390)" (#391" (#391)) @robertmaynard
• Add COMPONENT arguments for rapids_export to formatting file. (#389) @robertmaynard
• install_relocatable generate correct installed RESOURCE_SPEC_FILE (#386) @robertmaynard
• support_conda_env only add rpath-link flag to linkers that support it. (#384) @robertmaynard
• rapids_cpm_nvbench properly specify usage of external fmt library (#376) @robertmaynard
• rapids_cpm_spdlog properly specify usage of external fmt library (#375) @robertmaynard
• Patch nvbench to allow usage of external fmt (#373) @robertmaynard
• Support static builds of fmt (#372) @robertmaynard
• Update to latest nvbench (#371) @vyasr

📖 Documentation

• Fix misspelling of rapids_cpm_init (#385) @dagardner-nv

🚀 New Features

• rapids_test_install_relocatable tracks tests environment properties (#390) @robertmaynard
• rapids_test_install_relocatable EXCLUDE_FROM_ALL is now the default (#388) @robertmaynard
• Support downloading nvcomp CTK 11 or 12 binaries (#381) @robertmaynard
• Introduce clang-format to rapids-cmake to format C++ code examples (#378) @robertmaynard
• proprietary_binary now supports cuda toolkit version placeholders (#377) @robertmaynard
• Add rapids_test allowing projects to run gpu tests in parallel (#328) @robertmaynard
• Extend rapids_export to support the concept of optional COMPONENTS (#154) @robertmaynard

🛠️ Improvements

• Update to GCC 11 (#382) @bdice
• Make docs builds less verbose (#380) @AyodeAwe
• Update GHAs Workflows (#374) @ajschmidt8
• Use trap to handle errors in test scripts (#370) @AjayThorve
• Bump spdlog to 1.11, add fmt as dependency for spdlog (#368) @kkraus14
• Clean up and sort CPM packages. (#366) @bdice
• Update shared workflow branches (#365) @ajschmidt8
• Add fmt 9.1.0 (#364) @kkraus14
• Move date to build string in conda recipe (#359) @ajschmidt8
• Add docs build job (#347) @AyodeAwe

rapids-cmake 23.02.00 (9 Feb 2023)

🐛 Bug Fixes

• Remove incorrect deprecation for CMAKE_CUDA_ARCHITECTURES="NATIVE" (#355) @robertmaynard
• cpm: always_download now considers patches json entry (#353) @robertmaynard
• Use string literals for policy test messages so no escaping needed (#351) @robertmaynard
• Revert "Update spdlog to 1.11 ( latest version ) (#342)" (#346" (#346)) @bdice
• Revert update of libcudacxx 1.9 (#337) @robertmaynard
• rapids_cuda_patch_toolkit: Better handle non-standard toolkits (#324) @robertmaynard
• Revert "Upgrade spdlog to 1.10.0 (#312)" (#323" (#323)) @bdice
• rapids_cuda_init_architectures now supports CUDAARCHS env variable (#322) @robertmaynard
• Remove usage of FetchContent from tests to improve perf (#303) @robertmaynard

🚀 New Features

• Update nvCOMP version to 2.6.1 (#360) @vuule
• cpm: Rework always_download rules to be smarter (#348) @robertmaynard
• Add deprecation notice to passing "" to CMAKE_CUDA_ARCHITECTURES (#345) @robertmaynard
• Update to libcudacxx 1.9.1 to have a version >= CUDA Toolkit 12 (#343) @robertmaynard
• Update spdlog to 1.11 ( latest version ) (#342) @robertmaynard
• Update to nvcomp 2.6 (#341) @robertmaynard
• Add deprecation warnings for usage of ALL (#339) @robertmaynard
• rapids-cmake now errors out when CPM can't be downloaded (#335) @robertmaynard
• Update to nvcomp 2.5 (#333) @robertmaynard
• Update to libcudacxx 1.9 to match version found in CUDA Toolkit 12 (#332) @robertmaynard
• Update cuco git tag to fetch bug fixes and cleanups (#329) @PointKernel
• Fea/support cmake cuda architectures rapids value (#327) @robertmaynard
• Upgrade spdlog to 1.10.0 (#312) @kkraus14

🛠️ Improvements

• Update shared workflow branches (#361) @ajschmidt8
• Build against CUDA 11.8 (#344) @ajschmidt8
• Make generated find module targets global (#340) @vyasr
• Add codespell and whitespace linters to pre-commit hooks. (#338) @bdice
• Use pre-commit for style checks (#336) @bdice
• Branch 23.02 merge 22.12 (#331) @vyasr
• Update conda recipes. (#330) @bdice
• Fix typo. (#311) @vyasr

rapids-cmake 22.12.00 (8 Dec 2022)

🐛 Bug Fixes

• Don't use CMake 3.25.0 as it has a show stopping FindCUDAToolkit bug (#308) @robertmaynard
• Add missing CPM_ARGS to gbench (#294) @vyasr
• Patch results are only displayed once per invocation of CMake (#292) @robertmaynard
• Add thrust output iterator fix to rapids-cmake thrust patches (#291) @robertmaynard

📖 Documentation

• Update pull request template to match rest of RAPIDS (#280) @robertmaynard
• Clarify rapids_cuda_init_architectures behavior (#279) @robertmaynard

🚀 New Features

• Update cuco git tag (#302) @PointKernel
• Remove old CI files (#300) @robertmaynard
• Update cuco to version that supports Ada and Hopper (#299) @robertmaynard
• Move libcudacxx 1.8.1 so we support sm90 (#296) @robertmaynard
• Add ability to specify library directories for target rpaths (#295) @vyasr
• Add support for cloning Google benchmark (#293) @vyasr
• Add current_json_dir placeholder in json patch file values (#289) @robertmaynard
• Add sm90 ( Hopper ) to rapids-cmake "ALL" mode (#285) @robertmaynard
• Enable copy_prs ops-bot config (#284) @robertmaynard
• Add GitHub action workflow to rapids-cmake (#283) @robertmaynard
• Create conda package of patched dependencies (#275) @robertmaynard
• Switch thrust over to use rapids-cmake patches (#265) @robertmaynard

🛠️ Improvements

• Remove rapids-dependency-file-generator FIXME (#305) @ajschmidt8
• Add ninja as build dependency (#301) @ajschmidt8
• Forward merge 22.10 into 22.12 (#297) @vyasr

rapids-cmake 22.10.00 (12 Oct 2022)

🚨 Breaking Changes

• Update rapids-cmake to require cmake 3.23.1 (#227) @robertmaynard
• put $PREFIX before $BUILD_PREFIX in conda build (#182) @kkraus14

🐛 Bug Fixes

• Update to nvcomp 2.4.1 to fix zstd decompression (#286) @robertmaynard
• Restore rapids_cython_create_modules output variable name (#276) @robertmaynard
• rapids_cuda_init_architectures now obeys CUDAARCHS env variable (#270) @robertmaynard
• Update to Thrust 1.17.2 to fix cub ODR issues (#269) @robertmaynard
• conda_env: pass conda prefix as a rpath-link directory (#263) @robertmaynard
• Update cuCollections to fix issue with INSTALL_CUCO set to OFF. (#261) @bdice
• rapids_cpm_libcudacxx correct location of libcudacxx-config (#258) @robertmaynard
• Update rapids_find_generate_module to cmake 3.23 (#256) @robertmaynard
• Handle reconfiguring with USE_PROPRIETARY_BINARY value differing (#255) @robertmaynard
• rapids_cpm_thrust record build directory location of thrust-config (#254) @robertmaynard
• disable cuco install rules when no INSTALL_EXPORT_SET (#250) @robertmaynard
• Patch thrust and cub install rules to have proper header searches (#244) @robertmaynard
• Ensure that we install Thrust and Cub correctly. (#243) @robertmaynard
• Revert "Update to CPM v0.35.4 for URL downloads... (#236)" (#242) @robertmaynard
• put $PREFIX before $BUILD_PREFIX in conda build (#182) @kkraus14

📖 Documentation

• Correct broken patch_toolkit API docs, and CMake API cross references (#271) @robertmaynard
• Provide suggestions when encountering an incomplete GTest package (#247) @robertmaynard
• Docs: RAPIDS.cmake should be placed in current bin dir (#241) @robertmaynard
• Remove incorrect install location note on rapids_export (#232) @robertmaynard

🚀 New Features

• Update to CPM 0.35.6 as it has needed changes for cpm patching support. (#273) @robertmaynard
• Update to nvcomp 2.4 which now offers aarch64 binaries! (#272) @robertmaynard
• Support the concept of a patches to apply to a project built via CPM (#264) @robertmaynard
• Branch 22.10 merge 22.08 (#262) @robertmaynard
• Introduce rapids_cuda_patch_toolkit (#260) @robertmaynard
• Update libcudacxx to 1.8 (#253) @robertmaynard
• Update to CPM version 0.35.5 (#249) @robertmaynard
• Update to CPM v0.35.4 for URL downloads match the download time (#236) @robertmaynard
• rapids-cmake dependency tracking now understands COMPONENTS (#234) @robertmaynard
• Update to thrust 1.17 (#231) @robertmaynard
• Update to CPM v0.35.3 to support symlink build directories (#230) @robertmaynard
• Update rapids-cmake to require cmake 3.23.1 (#227) @robertmaynard
• Improve GPU detection by doing less subsequent executions (#222) @robertmaynard

🛠️ Improvements

• Fix typo in rapids-cmake-url (#267) @trxcllnt
• Ensure <pkg>_FOUND is set in the generated Find<pkg>.cmake file (#266) @trxcllnt
• Set CUDA_USE_STATIC_CUDA_RUNTIME to control legacy FindCUDA.cmakebehavior (#259) @trxcllnt
• Use the GitHub .zip URI instead of GIT_REPOSITORY and GIT_BRANCH (#257) @trxcllnt
• Update nvcomp to 2.3.3 (#221) @vyasr

rapids-cmake 22.08.00 (17 Aug 2022)

🐛 Bug Fixes

• json exclude flag behaves as expected libcudacx//thrust/nvcomp (#223) @robertmaynard
• Remove nvcomp dependency on CUDA::cudart_static (#218) @robertmaynard
• Timestamps for URL downloads match the download time (#215) @robertmaynard
• Revert "Update nvcomp to 2.3.2 (#209)" (#210" (#210)) @vyasr
• rapids-cmake won't ever use an existing variable starting with RAPIDS_ (#203) @robertmaynard

📖 Documentation

• Docs now provide rapids_find_package examples (#220) @robertmaynard
• Minor typo fix in api.rst (#207) @vyasr
• rapids_cpm_<pkgs> document handling of unparsed args (#206) @robertmaynard
• Docs/remove doc warnings (#205) @robertmaynard
• Fix docs: default behavior is to use a shallow git clone. (#204) @bdice
• Add rapids_cython to the html docs (#197) @robertmaynard

🚀 New Features

• More robust solution of CMake policy 135 (#224) @robertmaynard
• Update cuco git tag (#213) @PointKernel
• Revert "Revert "Update nvcomp to 2.3.2 (#209)" (#210)" (#211" (#210)" (#211)) @vyasr
• Update nvcomp to 2.3.2 (#209) @robertmaynard
• rapids_cpm_rmm no longer install when no INSTALL_EXPORT_SET listed (#202) @robertmaynard
• Adds support for pulling cuCollections using rapids-cmake (#201) @vyasr
• Add support for a prefix in Cython module targets (#198) @vyasr

🛠️ Improvements

• rapids_find_package() called with explicit version and REQUIRED should fail (#214) @trxcllnt

rapids-cmake 22.06.00 (7 June 2022)

🐛 Bug Fixes

• nvcomp install rules need to match the pre-built layout (#194) @robertmaynard
• Use target name variable. (#187) @bdice
• Remove unneeded message from rapids_export_package (#183) @robertmaynard
• rapids_cpm_thrust: Correctly find version 1.15.0 (#181) @robertmaynard
• rapids_cpm_thrust: Correctly find version 1.15.0 (#180) @robertmaynard

📖 Documentation

• Correct spelling mistake in cpm package docs (#188) @robertmaynard

🚀 New Features

• Add rapids_cpm_nvcomp with prebuilt binary support (#190) @robertmaynard
• Default Cython module RUNPATH to $ORIGIN and return the list of created targets (#189) @vyasr
• Add rapids-cython component for scikit-build based Python package builds (#184) @vyasr
• Add more exhaustive set of tests are version values of 0 (#178) @robertmaynard
• rapids_cpm_package_override now hooks into FetchContent (#164) @robertmaynard

🛠️ Improvements

• Update nvbench tag (#193) @PointKernel

rapids-cmake 22.04.00 (6 Apr 2022)

🐛 Bug Fixes

• rapids_export now handles explicit version values of 0 correctly (#174) @robertmaynard
• rapids_export now internally uses better named variables (#172) @robertmaynard
• rapids_cpm_gtest will properly find GTest 1.10 packages (#168) @robertmaynard
• CMAKE_CUDA_ARCHITECTURES ALL will not insert 62 or 72 (#161) @robertmaynard
• Tracked package versions are now not required, but preferred. (#160) @robertmaynard
• cpm_thrust would fail when provided only an install export set (#155) @robertmaynard
• rapids_export generated config.cmake no longer leaks variables (#149) @robertmaynard

📖 Documentation

• Docs use intersphinx correctly to link to CMake command docs (#159) @robertmaynard
• Example explains when you should use rapids_find_generate_module (#153) @robertmaynard
• Add CMake intersphinx support (#147) @bdice

🚀 New Features

• Bump CPM 0.35 for per package CPM_DOWNLOAD controls (#158) @robertmaynard
• Track package versions to the generated find_dependency calls (#156) @robertmaynard
• Update to latest nvbench (#150) @robertmaynard

🛠️ Improvements

• Temporarily disable new ops-bot functionality (#170) @ajschmidt8
• Use exact gtest version (#165) @trxcllnt
• Add .github/ops-bot.yaml config file (#163) @ajschmidt8

rapids-cmake 22.02.00 (2 Feb 2022)

🐛 Bug Fixes

• Ensure that nvbench doesn't require nvml when CUDA::nvml doesn't exist (#146) @robertmaynard
• rapids_cpm_libcudacxx handle CPM already finding libcudacxx before being called (#130) @robertmaynard

📖 Documentation

• Fix typos (#142) @ajschmidt8
• Fix type-o in docs <PackageName>_BINARY_DIR instead of <PackageName>_BINAR_DIR (#140) @dagardner-nv
• Set the always_download value in versions.json to the common case (#135) @robertmaynard
• Update Changelog to capture all 21.08 and 21.10 changes (#134) @robertmaynard
• Correct minor formatting issues (#132) @robertmaynard
• Document how to control the git rep/tag that RAPIDS.cmake uses (#131) @robertmaynard

🚀 New Features

• rapids-cmake now supports an empty package entry in the override file (#145) @robertmaynard
• Update NVBench for 22.02 to be the latest version (#144) @robertmaynard
• Update rapids-cmake packages to libcudacxx 1.7 (#143) @robertmaynard
• Update rapids-cmake packages to Thrust 1.15 (#138) @robertmaynard
• add exclude_from_all flag to version.json (#137) @robertmaynard
• Add PREFIX option to write_version_file / write_git_revision_file (#118) @robertmaynard

🛠️ Improvements

• Remove rapids_cmake_install_lib_dir unstable side effect checks (#136) @robertmaynard

rapids-cmake 21.12.00 (9 Dec 2021)

🐛 Bug Fixes

• rapids_cpm_libcudacxx install logic is safe for multiple inclusion (#124) @robertmaynard
• rapids_cpm_libcudacxx ensures CMAKE_INSTALL_INCLUDEDIR exists (#122) @robertmaynard
• rapids_cpm_find restores CPM variables when project was already added (#121) @robertmaynard
• rapids_cpm_thrust doesn't place temp file in a searched location (#120) @robertmaynard
• Require the exact version of Thrust in the versions.json file (#119) @trxcllnt
• CMake option second parameter is the help string, not the default value (#114) @robertmaynard
• Make sure we don't do a shallow clone on nvbench (#113) @robertmaynard
• Pin NVBench to a known working SHA1 (#112) @robertmaynard
• Build directory config.cmake now sets the correct targets to global (#110) @robertmaynard
• rapids_cpm_thrust installs to a location that won't be marked system (#98) @robertmaynard
• find_package now will find modules that CPM has downloaded. (#96) @robertmaynard
• rapids_cpm_thrust dont export namespaced thrust target (#93) @robertmaynard
• rapids_cpm_spdlog specifies the correct install variable (#91) @robertmaynard
• rapids_cpm_init: CPM_SOURCE_CACHE doesn't mean the CPM file exists (#87) @robertmaynard

📖 Documentation

• Better document that rapids_cpm_find supports arbitrary projects (#108) @robertmaynard
• Update the example to showcase rapids-cmake 21.12 (#107) @robertmaynard
• Properly generate rapids_cuda_init_runtime docs (#106) @robertmaynard

🚀 New Features

• Introduce rapids_cpm_libcudacxx (#111) @robertmaynard
• Record formatting rules for rapids_cpm_find DOWNLOAD_ONLY option (#94) @robertmaynard
• rapids_cmake_install_lib_dir now aware of GNUInstallDirs improvements in CMake 3.22 (#85) @robertmaynard
• rapids-cmake defaults to always download overridden packages (#83) @robertmaynard

🛠️ Improvements

• Prefer CPM_<pkg>_SOURCE dirs over find_package() in rapids_cpm_find (#92) @trxcllnt

rapids-cmake 21.10.00 (7 Oct 2021)

🐛 Bug Fixes

• Remove unneeded inclusions of the old setup_cpm_cache.cmake (#82) @robertmaynard
• Make sure rapids-cmake doesn't produce CMake syntax warnings (#80) @robertmaynard
• rapids_export verify DOCUMENTATION and FINAL_CODE_BLOCK exist (#75) @robertmaynard
• Make sure rapids_cpm_spdlog specifies the correct spdlog global targets (#71) @robertmaynard
• rapids_cpm_thrust specifies the correct install variable (#70) @robertmaynard
• FIX Install sphinxcontrib-moderncmakedomain in docs script (#69) @dillon-cullinan
• rapids_export_cpm(BUILD) captures location of locally found packages (#65) @robertmaynard
• Introduce rapids_cmake_install_lib_dir (#61) @robertmaynard
• rapids_export(BUILD) only creates alias targets to existing targets (#55) @robertmaynard
• rapids_find_package propagates variables from find_package (#54) @robertmaynard
• rapids_cpm_find is more invariant as one would expect (#51) @robertmaynard
• rapids-cmake tests properly state what C++ std levels they require (#46) @robertmaynard
• rapids-cmake always generates GLOBAL_TARGETS names correctly (#36) @robertmaynard

📖 Documentation

• Update update-version.sh (#84) @raydouglass
• Add rapids_export_find_package_root to api doc page (#76) @robertmaynard
• README.md now references online docs (#72) @robertmaynard
• Copyright year range now matches when rapids-cmake existed (#67) @robertmaynard
• cmake-format: Now aware of rapids_cmake_support_conda_env flags (#62) @robertmaynard
• Bug/correct invalid generate module doc layout (#47) @robertmaynard

🚀 New Features

• rapids-cmake SHOULD_FAIL tests verify the CMake Error string (#79) @robertmaynard
• Introduce rapids_cmake_write_git_revision_file (#77) @robertmaynard
• Allow projects to override version.json information (#74) @robertmaynard
• rapids_export_package(BUILD) captures location of locally found packages (#68) @robertmaynard
• Introduce rapids_export_find_package_root command (#64) @robertmaynard
• Introduce rapids_cpm_<preset> (#52) @robertmaynard
• Tests now can be SERIAL and use FetchContent to get rapids-cmake (#48) @robertmaynard
• rapids_export version support expanded to handle more use-cases (#37) @robertmaynard

🛠️ Improvements

• cpm tests now download less components and can be run in parallel. (#81) @robertmaynard
• Ensure that all rapids-cmake files have include guards (#63) @robertmaynard
• Introduce RAPIDS.cmake a better way to fetch rapids-cmake (#45) @robertmaynard
• ENH Replace gpuci_conda_retry with gpuci_mamba_retry (#44) @dillon-cullinan

rapids-cmake 21.08.00 (4 Aug 2021)

🚀 New Features

• Introduce rapids_cmake_write_version_file to generate a C++ version header (#23) @robertmaynard
• Introduce cmake-format-rapids-cmake to allow cmake-format to understand rapdids-cmake custom functions (#29) @robertmaynard

🛠️ Improvements

🐛 Bug Fixes

• ci/gpu/build.sh uses git tags to properly compute conda env (#43) @robertmaynard
• Make sure that rapids-cmake-dir cache variable is hidden (#40) @robertmaynard
• Correct regression specify rapids-cmake-dir as a cache variable (#39) @robertmaynard
• rapids-cmake add entries to CMAKE_MODULE_PATH on first config (#34) @robertmaynard
• Add tests that verify all paths in each rapids-.cmake file (#24) @robertmaynard
• Correct issue where rapids_export(DOCUMENTATION content was being ignored(#30) @robertmaynard
• rapids-cmake can now be correctly used by multiple adjacent directories (#33) @robertmaynard

rapids-cmake 21.06.00 (Date TBD)

Please see https://github.com/rapidsai/rapids-cmake/releases/tag/v21.06.0a for the latest changes to this development branch.

🚀 New Features

• Introduce rapids_cmake_parse_version for better version extraction (#20) @robertmaynard

🛠️ Improvements

• Verify that rapids-cmake always preserves CPM arguments (#18) @robertmaynard
• Add Sphinx based documentation for the project (#14) @robertmaynard
• rapids_export places the build export files in a location CPM can find. (#3) @robertmaynard

🐛 Bug Fixes

• Make sure we properly quote all CPM args (#17) @robertmaynard
• rapids_export correctly handles version strings with leading zeroes (#12) @robertmaynard
• rapids_export_write_language properly executes each time CMake is run (#10) @robertmaynard
• rapids_export properly sets version variables (#9) @robertmaynard
• rapids_export now obeys CMake config file naming convention (#8) @robertmaynard
• Refactor layout to enable adding CI and Documentation (#5) @robertmaynard