🚨 Breaking Changes

• Remove now unnecessary variable (#466) @vyasr

🐛 Bug Fixes

• Fix no-cuda build (#484) @madsbk
• Add license to kvikio wheel (#481) @raydouglass
• Fix release-after-main bug in get_primary_cuda_context (#472) @madsbk
• Intentionally leak static CUDA resources to avoid crash (part 2) (#462) @kingcrimsontianyu
• Fix several issues with wheel builds (#439) @KyleFromNVIDIA
• Fix missing 24.08 => 24.10 renames (#423) @madsbk

📖 Documentation

• Docs cleanup (#480) @madsbk

🚀 New Features

• Update nvcomp to 4.0.1 (#449) @vuule

🛠️ Improvements

• nvtx support macos (#474) @madsbk
• Update update-version.sh to use packaging lib (#469) @AyodeAwe
• Remove now unnecessary variable (#466) @vyasr
• Use CI workflow branch 'branch-24.10' again (#465) @jameslamb
• Update to flake8 7.1.1. (#463) @bdice
• remove pytest dependency in kvikio conda recipe (#461) @jameslamb
• Add support for Python 3.12, relax numcodecs pin (#458) @jameslamb
• Skip cufile and nvtx3 when cuda isn't found (#455) @madsbk
• Non-static bounce buffer option (#454) @madsbk
• Use nvcomp defaults for algo options. (#450) @bdice
• Update rapidsai/pre-commit-hooks (#448) @KyleFromNVIDIA
• Make the bounce buffer size configurable (#447) @madsbk
• Update build.sh (#445) @madsbk
• Drop Python 3.9 support (#443) @jameslamb
• Remove NumPy <2 pin (#441) @seberg
• C++ Test (#440) @madsbk
• config_path(): return a reference (#438) @madsbk
• Stream sync before calling cufile (#437) @madsbk
• Speeding up Python tests (#435) @madsbk
• Cython refactoring (#434) @madsbk
• Update pre-commit hooks (#433) @KyleFromNVIDIA
• Improve update-version.sh (#432) @bdice
• Use tool.scikit-build.cmake.version, set scikit-build-core minimum-version (#430) @jameslamb
• build.sh: --pydevelop (#429) @madsbk

🔗 Links

• Development Branch
• Compare with main branch

🚨 Breaking Changes

• Small improvements (#493) @kingcrimsontianyu

🐛 Bug Fixes

• Visibility of static class methods (#492) @madsbk

🛠️ Improvements

• make conda installs in CI stricter (#495) @jameslamb
• Small improvements (#493) @kingcrimsontianyu
• Prune workflows based on changed files (#489) @KyleFromNVIDIA
• Use nvcomp wheel instead of bundling nvcomp (#478) @KyleFromNVIDIA
• Touch ups to Array (#475) @jakirkham
• Remote IO: http support (#464) @madsbk

🐛 Bug Fixes

• fix compat_mode check typo (#421) @madsbk
• Fix O_DIRECT in defined (#416) @jakirkham
• cufile-wrapper: include the cuda-wrapper (#413) @madsbk
• Rename .devcontainers for CUDA 12.5 (#406) @jakirkham
• Use #ifdef to guard O_DIRECT usage (#404) @jakirkham
• Properly quote items in all (#399) @KyleFromNVIDIA

🛠️ Improvements

• Benchmarks clean up (#424) @madsbk
• split up CUDA-suffixed dependencies in dependencies.yaml (#411) @jameslamb
• Don't open with O_DIRECT when in compat mode (#410) @madsbk
• Fetch and use BS thread_pool (#408) @madsbk
• Use workflow branch 24.08 again (#407) @KyleFromNVIDIA
• Enable cuFile on ARM (#405) @jakirkham
• Build and test with CUDA 12.5.1 (#401) @KyleFromNVIDIA
• skip CMake 3.30.0 (#400) @jameslamb
• Use verify-alpha-spec hook (#398) @KyleFromNVIDIA
• Avoid --find-links in wheel jobs (#397) @jameslamb
• Adding NVTX annotations (#395) @madsbk
• Clean up error handling (#394) @madsbk
• Move content of the Python pkg to subdirectory python/kvikio/ (#392) @hcho3
• Adopt CI/packaging codeowners (#391) @bdice
• Remove text builds of documentation (#390) @vyasr
• use rapids-build-backend (#389) @jameslamb
• remove unnecessary 'setuptools' dependency (#385) @jameslamb
• add --rm and --name to devcontainer run args (#370) @trxcllnt
• add wheel output (#369) @msarahan

🐛 Bug Fixes

• Fix conda recipe for kvikio (#381) @KyleFromNVIDIA
• Move cuFile linking to kvikio target (#379) @KyleFromNVIDIA

🛠️ Improvements

• Initial Python Interface for cufile Async IO (#376) @K-Wu
• Enable Warnings as errors for Python tests (#375) @mroeschke
• POSIX IO: use async memcpy (#374) @madsbk
• No-CUDA build (#373) @madsbk
• Refactor: move cufile header checks to the shim layer (#372) @madsbk
• Migrate to {{ stdlib("c") }} (#371) @hcho3
• Add deflate codec (#364) @ap--

🚨 Breaking Changes

• remove legate (#345) @madsbk

🐛 Bug Fixes

• Update pre-commit-hooks to v0.0.3 (#354) @KyleFromNVIDIA
• Defer cuFile feature checks until finding kvikio package (#342) @bdice
• handle more RAPIDS version formats in update-version.sh (#331) @jameslamb

🚀 New Features

• Support CUDA 12.2 (#328) @jameslamb

🛠️ Improvements

• doc: cuda toolkit 12.0 => 12.2 (#360) @madsbk
• Use conda env create --yes instead of --force (#357) @bdice
• Prevent path conflict in builds (#356) @AyodeAwe
• Add upper bound to prevent usage of NumPy 2 (#353) @bdice
• Replace local copyright check with pre-commit-hooks verify-copyright (#352) @KyleFromNVIDIA
• Remove hard-coding of RAPIDS version where possible (#351) @KyleFromNVIDIA
• Add support for Python 3.11, require NumPy 1.23+ (#350) @jameslamb
• remove legate (#345) @madsbk
• Update devcontainers to CUDA Toolkit 12.2 (#344) @trxcllnt
• target branch-24.04 for GitHub Actions workflows (#343) @jameslamb
• Add environment-agnostic scripts for running ctests and pytests (#338) @trxcllnt
• Update ops-bot.yaml (#335) @AyodeAwe

🚨 Breaking Changes

• Switch to scikit-build-core (#325) @vyasr

🐛 Bug Fixes

• Bump to nvcomp 3.0.6. (#346) @bdice
• Exclude tests from builds (#336) @vyasr
• Update build.sh (#332) @madsbk

🛠️ Improvements

• Remove usages of rapids-env-update (#329) @KyleFromNVIDIA
• refactor CUDA versions in dependencies.yaml (#327) @jameslamb
• Switch to scikit-build-core (#325) @vyasr
• Update nvcomp (#324) @vyasr
• Add timer to basic_io example (#321) @yncxcw
• Forward-merge branch-23.12 to branch-24.02 (#318) @bdice
• Re-enable devcontainer CI. (#285) @trxcllnt

🚨 Breaking Changes

• Switch to scikit-build-core (#325) @vyasr

🐛 Bug Fixes

• Exclude tests from builds (#336) @vyasr
• Update build.sh (#332) @madsbk

🛠️ Improvements

• Remove usages of rapids-env-update (#329) @KyleFromNVIDIA
• refactor CUDA versions in dependencies.yaml (#327) @jameslamb
• Switch to scikit-build-core (#325) @vyasr
• Update nvcomp (#324) @vyasr
• Add timer to basic_io example (#321) @yncxcw
• Forward-merge branch-23.12 to branch-24.02 (#318) @bdice
• Re-enable devcontainer CI. (#285) @trxcllnt

🚨 Breaking Changes

• Update nvcomp to 3.0.4 (includes API changes) (#314) @vuule

🐛 Bug Fixes

• Remove duplicated thread-pool API (#308) @madsbk
• updated the nvcomp notebook to use the new API (#294) @madsbk

🚀 New Features

• Update nvcomp to 3.0.4 (includes API changes) (#314) @vuule

🛠️ Improvements

• Build concurrency for nightly and merge triggers (#319) @bdice
• Revert rapids-cmake branch. (#316) @bdice
• Support no compressor in open_cupy_array() (#312) @madsbk
• Update shared-action-workflows references (#305) @AyodeAwe
• Use branch-23.12 workflows. (#304) @bdice
• Update rapids-cmake functions to non-deprecated signatures (#301) @robertmaynard
• Unify the CUDA Codecs (#298) @madsbk
• Improve performance of nvCOMP batch codec. (#293) @Alexey-Kamenev
• Merge branch-23.10 into branch-23.12 and fix devcontainer CI workflow. (#292) @bdice
• kvikio: Build CUDA 12.0 ARM conda packages. (#282) @bdice

🚨 Breaking Changes

• Update to Cython 3.0.0 (#258) @vyasr

🐛 Bug Fixes

• Add numcodecs pin (#300) @vyasr
• Add missed filename to sed_runner call (#286) @raydouglass
• Use conda mambabuild not mamba mambabuild (#278) @bdice
• fixes #254 (#262) @madsbk

📖 Documentation

• minor doc fixes (#279) @madsbk
• Docs (#268) @madsbk
• Zarr notebook (#261) @madsbk

🛠️ Improvements

• Use branch-23.10 for devcontainers workflow. (#289) @bdice
• Update image names (#284) @AyodeAwe
• Update to clang 16.0.6. (#280) @bdice
• Update doxygen to 1.9.1 (#277) @vyasr
• Async I/O using by-value arguments (#275) @madsbk
• Zarr-IO Benchmark (#274) @madsbk
• Add KvikIO devcontainers (#273) @trxcllnt
• async: fall back to blocking (#272) @madsbk
• Unify batch and stream API check (#271) @madsbk
• Use copy-pr-bot (#269) @ajschmidt8
• Zarr+CuPy+GDS+nvCOMP made easy (#267) @madsbk
• Remove sphinx pinning (#260) @vyasr
• Initial changes to support cufile stream I/O. (#259) @tell-rebanta
• Update to Cython 3.0.0 (#258) @vyasr
• Modernize Python build (#257) @vyasr
• Enable roundtrip for nvCOMP batch codecs. (#253) @Alexey-Kamenev

🐛 Bug Fixes

• Add nvcomp support to older CUDA 11 versions on aarch64. (#255) @bdice
• Unify KVIKIO_CUFILE_FOUND (#243) @madsbk
• Disable the batch API when in compatibility mode (#239) @madsbk
• Fix libcufile dependency. (#237) @bdice

🚀 New Features

• KvikIO: Build CUDA 12 packages (#224) @bdice

🛠️ Improvements

• Revert CUDA 12.0 CI workflows to branch-23.08. (#252) @bdice
• Add support for nvCOMP batch API (#249) @Alexey-Kamenev
• Use cuda-version to constrain cudatoolkit. (#247) @bdice
• Make C++ & Python teams owners of legate/ (#246) @jakirkham
• Use nvcomp conda package. (#245) @bdice
• Adding code owners (#244) @madsbk
• Clean up dependency lists (#241) @vyasr
• Clean up isort configs (#240) @vyasr
• Update to CMake 3.26.4 (#238) @vyasr
• use rapids-upload-docs script (#234) @AyodeAwe
• Migrate as much as possible to pyproject.toml, stop using versioneer to manage versions, update dependencies.yaml. (#232) @bdice
• Remove documentation build scripts for Jenkins (#230) @ajschmidt8
• Use the Zarr's new getitems() API (#131) @madsbk