rmm

🐛 Bug Fixes

• Clarifies Python requirements and version constraints (#1037) @jakirkham
• Use lib (not lib64) for libraries (#1024) @jakirkham
• Properly enable Cython docstrings. (#1020) @vyasr
• Update RMMNumbaManager to handle NUMBA_CUDA_USE_NVIDIA_BINDING=1 (#1004) @brandon-b-miller

📖 Documentation

• Clarify using RMM with other Python libraries (#1034) @jrhemstad
• Replace to_device with DeviceBuffer.to_device (#1033) @wence-
• Documentation Fix: Replace cudf::logic_error with rmm::logic_error (#1021) @codereport

🚀 New Features

• Add rmm::exec_policy_nosync (#1009) @fkallen
• Callback memory resource (#980) @shwina

🛠️ Improvements

• Fix conda recipes for conda compilers (#1043) @Ethyling
• Use new rapids-cython component of rapids-cmake to simplify builds (#1031) @vyasr
• Merge branch-22.04 to branch-22.06 (#1028) @jakirkham
• Update CMake pinning to just avoid 3.23.0. (#1023) @vyasr
• Build python using conda in GPU jobs (#1017) @Ethyling
• Remove pip requirements file. (#1015) @bdice
• Clean up Thrust includes. (#1011) @bdice
• Update black version (#1010) @vyasr
• Update cmake-format version for pre-commit and environments. (#995) @vyasr
• Use conda compilers (#977) @Ethyling
• Build conda packages using mambabuild (#900) @Ethyling

v22.04.01

🐛 Bug Fixes

• Add cuda-python dependency to pyproject.toml (#994) @sevagh
• Disable opportunistic reuse in async mr when cuda driver < 11.5 (#993) @rongou
• Use CUDA 11.2+ features via dlopen (#990) @robertmaynard
• Skip async mr tests when cuda runtime/driver < 11.2 (#986) @rongou
• Fix warning/error in debug assertion in device_uvector.hpp (#979) @harrism
• Fix signed/unsigned comparison warning (#970) @jlowe
• Fix comparison of async MRs with different underlying pools. (#965) @harrism

🚀 New Features

• Use scikit-build for the build process (#976) @vyasr

🛠️ Improvements

• Temporarily disable new ops-bot functionality (#1005) @ajschmidt8
• Rename librmm_tests to librmm-tests (#1000) @ajschmidt8
• Update librmm conda recipe (#997) @ajschmidt8
• Remove no_cma/has_cma variants (#996) @ajschmidt8
• Fix free-before-alloc in multithreaded test (#992) @aladram
• Add .github/ops-bot.yaml config file (#991) @ajschmidt8
• Log allocation failures (#988) @rongou
• Update librmm conda outputs (#983) @ajschmidt8
• Bump Python requirements in setup.cfg and rmm_dev.yml (#982) @shwina
• New benchmark compares concurrent throughput of device_vector and device_uvector (#981) @harrism
• Update librmm recipe to output librmm_tests package (#978) @ajschmidt8
• Update upload.sh to use --croot (#975) @AyodeAwe
• Fix conda uploads (#974) @ajschmidt8
• Add CMake install rules for tests (#969) @ajschmidt8
• Add device_buffer::ssize() and device_uvector::ssize() (#966) @harrism
• Added yml file for cudatoolkit version 11.6 (#964) @alhad-deshpande
• Replace ccache with sccache (#963) @ajschmidt8
• Make pool_memory_resource::pool_size() public (#962) @shwina
• Allow construction of cuda_async_memory_resource from existing pool (#889) @fkallen

🐛 Bug Fixes

• Use numba to get CUDA runtime version. (#946) @bdice
• Temporarily disable warnings for unknown pragmas (#942) @harrism
• Build benchmarks in RMM CI (#941) @harrism
• Headers that use std::thread now include <thread> (#938) @robertmaynard
• Fix failing stream test with a debug-only death test (#934) @harrism
• Prevent DeviceBuffer DeviceMemoryResource premature release (#931) @viclafargue
• Fix failing tracking test (#929) @harrism

🛠️ Improvements

• Prepare upload scripts for Python 3.7 removal (#952) @Ethyling
• Fix imports tests syntax (#935) @Ethyling
• Remove IncludeCategories from .clang-format (#933) @codereport
• Replace use of custom CUDA bindings with CUDA-Python (#930) @shwina
• Remove setup.py from update-release.sh script (#926) @ajschmidt8
• Improve C++ Test Coverage (#920) @harrism
• Improve the Arena allocator to reduce memory fragmentation (#916) @rongou
• Simplify CMake linting with cmake-format (#913) @vyasr

🚨 Breaking Changes

• Parameterize exception type caught by failure_callback_resource_adaptor (#898) @harrism

🐛 Bug Fixes

• Update recipes for Enhanced Compatibility (#910) @ajschmidt8
• Fix librmm uploads (#909) @ajschmidt8
• Use spdlog/fmt/ostr.h as it supports external fmt library (#907) @robertmaynard
• Fix variable names in logging macro calls (#897) @harrism
• Keep rapids cmake version in sync (#876) @robertmaynard

📖 Documentation

• Replace to_device() in docs with DeviceBuffer.to_device() (#902) @shwina
• Fix return value docs for supports_get_mem_info (#884) @harrism

🚀 New Features

• Out-of-memory callback resource adaptor (#892) @madsbk

🛠️ Improvements

• suppress spurious clang-tidy warnings in debug macros (#914) @rongou
• C++ code coverage support (#905) @harrism
• Provide ./build.sh flag to control CUDA async malloc support (#901) @robertmaynard
• Parameterize exception type caught by failure_callback_resource_adaptor (#898) @harrism
• Throw rmm::out_of_memory when we know for sure (#894) @rongou
• Update conda recipes for Enhanced Compatibility effort (#893) @ajschmidt8
• Add functions to query the stream of device_uvector and device_scalar (#887) @fkallen
• Add spdlog to install export set (#886) @trxcllnt

v21.10.01

🚨 Breaking Changes

• Delete cuda_async_memory_resource copy/move ctors/operators (#860) @jrhemstad

🐛 Bug Fixes

• Fix parameter name in asserts (#875) @vyasr
• Disallow zero-size stream pools (#873) @harrism
• Correct namespace usage in host memory resources (#872) @divyegala
• fix race condition in limiting resource adapter (#869) @rongou
• Install the right cudatoolkit in the conda env in gpu/build.sh (#864) @shwina
• Disable copy/move ctors and operator= from free_list classes (#862) @harrism
• Delete cuda_async_memory_resource copy/move ctors/operators (#860) @jrhemstad
• Improve concurrency of stream_ordered_memory_resource by stealing less (#851) @harrism
• Use the new RAPIDS.cmake to fetch rapids-cmake (#838) @robertmaynard

📖 Documentation

• Forward-merge branch-21.08 to branch-21.10 (#846) @jakirkham

🛠️ Improvements

• Forward-merge branch-21.08 into branch-21.10 (#877) @ajschmidt8
• Add .clang-tidy and fix clang-tidy warnings (#857) @harrism
• Update to use rapids-cmake 21.10 pre-configured packages (#854) @robertmaynard
• Clean up: use std::size_t, include cstddef and aligned.hpp where missing (#852) @harrism
• tweak the arena mr to reduce fragmentation (#845) @rongou
• Fix transitive include in cuda_device header (#843) @wphicks
• Refactor cmake style (#842) @robertmaynard
• add multi stream allocations benchmark. (#841) @cwharris
• Enforce default visibility for get_map. (#833) @trivialfis
• ENH Replace gpuci_conda_retry with gpuci_mamba_retry (#823) @dillon-cullinan
• Execution policy class (#816) @viclafargue

v21.08.02

v21.08.01

🚨 Breaking Changes

• Refactor rmm::device_scalar in terms of rmm::device_uvector (#789) @harrism
• Explicit streams in device_buffer (#775) @harrism

🐛 Bug Fixes

• Pin spdlog in dev conda envs (#835) @trxcllnt
• Pinning spdlog because recent updates are causing compile issues. (#831) @cjnolet
• update isort to 5.6.4 (#822) @cwharris
• fix align_up namespace in aligned_resource_adaptor.hpp (#820) @rongou
• Run updated isort hook on pxd files (#812) @charlesbluca
• find_package(RMM) can now be called multiple times safely (#811) @robertmaynard
• Fix building on CUDA 11.3 (#809) @benfred
• Remove leading zeros in version_config.hpp (#793) @hcho3

📖 Documentation

• Fix PoolMemoryResource Python doc examples (#807) @harrism
• Fix incorrect href in README.md (#804) @benchislett
• Update build instruction in README (#797) @hcho3
• Document compute sanitizer memcheck support (#790) @harrism

🚀 New Features

• Bump isort, enable Cython package resorting (#806) @charlesbluca
• Support multiple output sinks in logging_resource_adaptor (#791) @harrism
• Add Statistics Resource Adaptor and cython bindings to tracking_resource_adaptor and statistics_resource_adaptor (#626) @mdemoret-nv

🛠️ Improvements

• Fix isort in cuda_stream_view.pxd (#827) @harrism
• Cython extension for rmm::cuda_stream_pool (#818) @divyegala
• Fix building on cuda 11.4 (#817) @benfred
• Updating Clang Version to 11.0.0 (#814) @codereport
• Add spdlog to rmm-exports if found by CPM (#810) @trxcllnt
• Fix 21.08 forward-merge conflicts (#803) @ajschmidt8
• RMM now leverages rapids-cmake to reduce CMake boilerplate (#800) @robertmaynard
• Refactor rmm::device_scalar in terms of rmm::device_uvector (#789) @harrism
• make it easier to include rmm in other projects (#788) @rongou
• Compile Cython with C++17. (#787) @vyasr
• Fix Merge Conflicts (#786) @ajschmidt8
• Explicit streams in device_buffer (#775) @harrism

🐛 Bug Fixes

• FindThrust now guards against multiple inclusion by different consumers (#784) @robertmaynard

📖 Documentation

• Document synchronization requirements on device_buffer copy ctors (#772) @harrism

🚀 New Features

• add a resource adapter to align on a specified size (#768) @rongou

🛠️ Improvements

• Update environment variable used to determine cuda_version (#785) @ajschmidt8
• Update CHANGELOG.md links for calver (#781) @ajschmidt8
• Merge branch-0.19 into branch-21.06 (#779) @ajschmidt8
• Update docs build script (#776) @ajschmidt8
• upgrade spdlog to 1.8.5 (#658) @rongou

🚨 Breaking Changes

• Avoid potential race conditions in device_scalar/device_uvector setters (#725) @harrism

🐛 Bug Fixes

• Fix typo in setup.py (#746) @galipremsagar
• Revert "Update rmm conda recipe pinning of librmm" (#743) @raydouglass
• Update rmm conda recipe pinning of librmm (#738) @mike-wendt
• RMM doesn't require the CUDA language to be enabled by consumers (#737) @robertmaynard
• Fix setup.py to work in a non-conda environment setup (#733) @galipremsagar
• Fix auto-detecting GPU architectures (#727) @trxcllnt
• CMAKE_CUDA_ARCHITECTURES doesn't change when build-system invokes cmake (#726) @robertmaynard
• Ship memory_resource_wrappers.hpp as package_data (#715) @shwina
• Only include SetGPUArchs in the top-level CMakeLists.txt (#713) @trxcllnt
• Fix unknown CMake command "CPMFindPackage" (#699) @standbyme

📖 Documentation

• Fix host_memory_resource signature typo (#728) @miguelusque

🚀 New Features

• Clarify log file name behaviour in docs (#722) @shwina
• Add Cython definitions for device_uvector (#720) @shwina
• Python bindings for cuda_async_memory_resource (#718) @shwina

🛠️ Improvements

• Fix cython tests (#749) @galipremsagar
• Add requirements for rmm (#739) @galipremsagar
• device_uvector can be used within thrust::optional (#734) @robertmaynard
• arena_memory_resource optimization: disable tracking allocated blocks by default (#732) @rongou
• Remove CMAKE_CURRENT_BINARY_DIR path in rmm's target_include_directories (#731) @trxcllnt
• set CMAKE_CUDA_ARCHITECTURES to OFF instead of undefined (#729) @trxcllnt
• Avoid potential race conditions in device_scalar/device_uvector setters (#725) @harrism
• Update Changelog Link (#723) @ajschmidt8
• Prepare Changelog for Automation (#717) @ajschmidt8
• Update 0.18 changelog entry (#716) @ajschmidt8
• Simplify cmake cuda architectures handling (#709) @robertmaynard
• Build only compute for the newest arch in CMAKE_CUDA_ARCHITECTURES (#706) @robertmaynard
• ENH Build with Ninja & Pass ccache variables to conda recipe (#705) @dillon-cullinan
• pool_memory_resource optimization: disable tracking allocated blocks by default (#702) @harrism
• Allow the build directory of rmm to be used for find_package(rmm) (#698) @robertmaynard
• Adds a linear accessor to RMM cuda stream pool (#696) @afender
• Fix merge conflicts for #692 (#694) @ajschmidt8
• Fix merge conflicts for #692 (#693) @ajschmidt8
• Remove C++ Wrappers in memory_resource_adaptors.hpp Needed by Cython (#662) @mdemoret-nv
• Improve Cython Lifetime Management by Adding References in DeviceBuffer (#661) @mdemoret-nv
• Add support for streams in CuPy allocator (#654) @pentschev

Breaking Changes 🚨

• Remove DeviceBuffer synchronization on default stream (#650) @pentschev
• Add a Stream class that wraps CuPy/Numba/CudaStream (#636) @shwina

Bug Fixes 🐛

• SetGPUArchs updated to work around a CMake FindCUDAToolkit issue (#695) @robertmaynard
• Remove duplicate conda build command (#670) @raydouglass
• Update CMakeLists.txt VERSION to 0.18.0 (#665) @trxcllnt
• Fix wrong attribute names leading to DEBUG log build issues (#653) @pentschev

Documentation 📖

• Correct inconsistencies in README and CONTRIBUTING docs (#682) @robertmaynard
• Enable tag generation for doxygen (#672) @ajschmidt8
• Document that managed_memory_resource does not work with NVIDIA vGPU (#656) @harrism

New Features 🚀

• Enabling/disabling logging after initialization (#678) @shwina
• cuda_async_memory_resource built on cudaMallocAsync (#676) @harrism
• Create labeler.yml (#669) @jolorunyomi
• Expose the version string in C++ and Python (#666) @hcho3
• Add a CUDA stream pool (#659) @harrism
• Add a Stream class that wraps CuPy/Numba/CudaStream (#636) @shwina

Improvements 🛠️

• Update stale GHA with exemptions & new labels (#707) @mike-wendt
• Add GHA to mark issues/prs as stale/rotten (#700) @Ethyling
• Auto-label PRs based on their content (#691) @ajschmidt8
• Prepare Changelog for Automation (#688) @ajschmidt8
• Build.sh use cmake --build to drive build system invocation (#686) @robertmaynard
• Fix failed automerge (#683) @harrism
• Auto-label PRs based on their content (#681) @jolorunyomi
• Build RMM tests/benchmarks with -Wall flag (#674) @trxcllnt
• Remove DeviceBuffer synchronization on default stream (#650) @pentschev
• Simplify rmm::exec_policy and refactor Thrust support (#647) @harrism

v0.17.0 Release

v0.16.0 Release

v0.15.0 Release

The second release of RMM as a standalone repository. Previously it was part of https://github.com/rapidsai/cudf This release matches the version numbering of cuDF.

The initial release of RMM as a standalone repository. Previously it was part of https://github.com/rapidsai/cudf This release starts from the version number of cuDF at the time RMM was split.