|
1 | | -# 0.18.0 |
| 1 | +# RMM 0.18.0 (24 Feb 2021) |
2 | 2 |
|
3 | | -Please see https://github.com/rapidsai/rmm/releases/tag/branch-0.18-latest for the latest changes to this development branch. |
| 3 | +## Breaking Changes 🚨 |
| 4 | + |
| 5 | +- Remove DeviceBuffer synchronization on default stream (#650) @pentschev |
| 6 | +- Add a Stream class that wraps CuPy/Numba/CudaStream (#636) @shwina |
| 7 | + |
| 8 | +## Bug Fixes 🐛 |
| 9 | + |
| 10 | +- SetGPUArchs updated to work around a CMake FindCUDAToolkit issue (#695) @robertmaynard |
| 11 | +- Remove duplicate conda build command (#670) @raydouglass |
| 12 | +- Update CMakeLists.txt VERSION to 0.18.0 (#665) @trxcllnt |
| 13 | +- Fix wrong attribute names leading to DEBUG log build issues (#653) @pentschev |
| 14 | + |
| 15 | +## Documentation 📖 |
| 16 | + |
| 17 | +- Correct inconsistencies in README and CONTRIBUTING docs (#682) @robertmaynard |
| 18 | +- Enable tag generation for doxygen (#672) @ajschmidt8 |
| 19 | +- Document that `managed_memory_resource` does not work with NVIDIA vGPU (#656) @harrism |
| 20 | + |
| 21 | +## New Features 🚀 |
| 22 | + |
| 23 | +- Enabling/disabling logging after initialization (#678) @shwina |
| 24 | +- `cuda_async_memory_resource` built on `cudaMallocAsync` (#676) @harrism |
| 25 | +- Create labeler.yml (#669) @jolorunyomi |
| 26 | +- Expose the version string in C++ and Python (#666) @hcho3 |
| 27 | +- Add a CUDA stream pool (#659) @harrism |
| 28 | +- Add a Stream class that wraps CuPy/Numba/CudaStream (#636) @shwina |
| 29 | + |
| 30 | +## Improvements 🛠️ |
| 31 | + |
| 32 | +- Update stale GHA with exemptions & new labels (#707) @mike-wendt |
| 33 | +- Add GHA to mark issues/prs as stale/rotten (#700) @Ethyling |
| 34 | +- Auto-label PRs based on their content (#691) @ajschmidt8 |
| 35 | +- Prepare Changelog for Automation (#688) @ajschmidt8 |
| 36 | +- Build.sh use cmake --build to drive build system invocation (#686) @robertmaynard |
| 37 | +- Fix failed automerge (#683) @harrism |
| 38 | +- Auto-label PRs based on their content (#681) @jolorunyomi |
| 39 | +- Build RMM tests/benchmarks with -Wall flag (#674) @trxcllnt |
| 40 | +- Remove DeviceBuffer synchronization on default stream (#650) @pentschev |
| 41 | +- Simplify `rmm::exec_policy` and refactor Thrust support (#647) @harrism |
4 | 42 |
|
5 | 43 | # RMM 0.17.0 (10 Dec 2020) |
6 | 44 |
|
|
0 commit comments