Skip to content

Latest commit

 

History

History
99 lines (72 loc) · 4.01 KB

RELEASE_NOTES.md

File metadata and controls

99 lines (72 loc) · 4.01 KB
Raw

CARE Software Release Notes

Notes describing significant changes in each CARE release are documented in this file.

The format of this file is based on Keep a Changelog.

[Unreleased] - Release date YYYY-MM-DD

Added

  • Added CARE_DEEP_COPY_RAW_PTR configuration option.
  • Added ATOMIC_SUB, ATOMIC_LOAD, ATOMIC_STORE, ATOMIC_EXCHANGE, and ATOMIC_CAS macros.

Removed

  • Removed Accessor template parameter from host_device_ptr.
  • Removed NoOpAccessor and RaceConditionAccessor. It is recommended to use ThreadSanitizer (TSAN) instead to locate race conditions.
  • Removed CARE_ENABLE_RACE_DETECTION configuration option.
  • Removed implicit conversions between raw pointers and host_device_ptrs/host_ptrs and the corresponding CARE_ENABLE_IMPLICIT_CONVERSIONS configuration option.

Changed

  • Renamed host_device_ptr::getPointer to host_device_ptr::data.

Fixed

  • Replaced calls to chai::ManagedArray::getPointer (previously deprecated and now removed) with calls to chai::ManagedArray::data.

[Version 0.14.1] - Release date 2024-10-15

Fixed

  • Explicitly define host_device_map constructors since some versions of CUDA do not properly generate them.

[Version 0.14.0] - Release date 2024-09-11

Added

  • Added default and move constructors and move assignment operator to host_device_map

Changed

  • Default policies concerning reductions were updated to RAJA's newly recommended policies.
  • Now using '<' in care::BinarySearch.

Fixed

  • Removed C++17 features so that CARE is C++14 compliant. The next release of CARE will require C++17.
  • LLNL_GlobalID is no longer required downstream from CARE if it is disabled in CARE.
  • Const correctness fix in uniqArray API.
  • Sequential IntersectArrays now allocates data in a way that is consistent with the memory model CHAI uses for Hip GPU builds.

[Version 0.13.3] - Release date 2024-07-31

Fixed

  • Replaced loop_work alias with seq_work (loop_work was removed in RAJA v2024.02.2)
  • Fixed CHUNKED loop macro implementations

[Version 0.13.2] - Release date 2024-07-29

Changed

  • Updated to Umpire/RAJA/CHAI v2024.07.0
  • Updated minimum required CMake to 3.23

[Version 0.13.1] - Release date 2024-06-27

Changed

  • Updated to CHAI v2024.02.2

[Version 0.13.0] - Release date 2024-06-11

Added

  • Alias for execution policy specificially for kernels with reductions
  • Chunked loop policies

[Version 0.12.0] - Release date 2024-03-11

Added

  • Support for APUs with a single memory space.
  • ArrayDup overloads
  • LocalSortPairs for a thread local simultaneous sort
  • Better support for unsigned and 64 bit integers (explicit instantiations of some algorithms, a SCAN_LOOP_64 macro)

Removed

  • Camp submodule
  • radiuss-ci submodule
  • The chai_force_sync function used for debugging (the corresponding functionality in CHAI has also been removed)

Changed

  • When building with submodules, they now need to be initialized recursively
  • When building with external libraries, specify the install location with -D<uppercase name>_DIR (previously, the lowercase version was also accepted)
  • Only tests are built by default now. Docs, examples, and benchmarks must be enabled explicitly.
  • The CARE_ENABLE_* options for tests, docs, examples, and benchmarks now CMake dependent options based on the corresponding ENABLE_* options
  • care-config.cmake has been moved to ${INSTALL_PREFIX}/lib/cmake/care and now properly exports CMake targets for care

Fixed

  • Eliminated some unnecessary data motion for GPU builds
  • Several fixes were added for shared library builds on Windows
  • Some warnings have been fixed