From 6e8b9788f6a9bdf9921f403f5a7125b57097f9ee Mon Sep 17 00:00:00 2001 From: AJ Schmidt Date: Thu, 20 May 2021 15:52:20 -0400 Subject: [PATCH] update changelog (#777) --- CHANGELOG.md | 55 +++++++++++++++++++++++++++++++++++++++++++++++++--- 1 file changed, 52 insertions(+), 3 deletions(-) diff --git a/CHANGELOG.md b/CHANGELOG.md index ca0aa47a2..b72b1e10a 100644 --- a/CHANGELOG.md +++ b/CHANGELOG.md @@ -1,6 +1,55 @@ -# RMM 0.19.0 (Date TBD) - -Please see https://github.com/rapidsai/rmm/releases/tag/v0.19.0a for the latest changes to this development branch. +# RMM 0.19.0 (21 Apr 2021) + +## 🚨 Breaking Changes + +- Avoid potential race conditions in device_scalar/device_uvector setters ([#725](https://github.com/rapidsai/rmm/pull/725)) [@harrism](https://github.com/harrism) + +## 🐛 Bug Fixes + +- Fix typo in setup.py ([#746](https://github.com/rapidsai/rmm/pull/746)) [@galipremsagar](https://github.com/galipremsagar) +- Revert "Update `rmm` conda recipe pinning of `librmm`" ([#743](https://github.com/rapidsai/rmm/pull/743)) [@raydouglass](https://github.com/raydouglass) +- Update `rmm` conda recipe pinning of `librmm` ([#738](https://github.com/rapidsai/rmm/pull/738)) [@mike-wendt](https://github.com/mike-wendt) +- RMM doesn't require the CUDA language to be enabled by consumers ([#737](https://github.com/rapidsai/rmm/pull/737)) [@robertmaynard](https://github.com/robertmaynard) +- Fix setup.py to work in a non-conda environment setup ([#733](https://github.com/rapidsai/rmm/pull/733)) [@galipremsagar](https://github.com/galipremsagar) +- Fix auto-detecting GPU architectures ([#727](https://github.com/rapidsai/rmm/pull/727)) [@trxcllnt](https://github.com/trxcllnt) +- CMAKE_CUDA_ARCHITECTURES doesn't change when build-system invokes cmake ([#726](https://github.com/rapidsai/rmm/pull/726)) [@robertmaynard](https://github.com/robertmaynard) +- Ship memory_resource_wrappers.hpp as package_data ([#715](https://github.com/rapidsai/rmm/pull/715)) [@shwina](https://github.com/shwina) +- Only include SetGPUArchs in the top-level CMakeLists.txt ([#713](https://github.com/rapidsai/rmm/pull/713)) [@trxcllnt](https://github.com/trxcllnt) +- Fix unknown CMake command "CPMFindPackage" ([#699](https://github.com/rapidsai/rmm/pull/699)) [@standbyme](https://github.com/standbyme) + +## 📖 Documentation + +- Fix host_memory_resource signature typo ([#728](https://github.com/rapidsai/rmm/pull/728)) [@miguelusque](https://github.com/miguelusque) + +## 🚀 New Features + +- Clarify log file name behaviour in docs ([#722](https://github.com/rapidsai/rmm/pull/722)) [@shwina](https://github.com/shwina) +- Add Cython definitions for device_uvector ([#720](https://github.com/rapidsai/rmm/pull/720)) [@shwina](https://github.com/shwina) +- Python bindings for `cuda_async_memory_resource` ([#718](https://github.com/rapidsai/rmm/pull/718)) [@shwina](https://github.com/shwina) + +## 🛠️ Improvements + +- Fix cython tests ([#749](https://github.com/rapidsai/rmm/pull/749)) [@galipremsagar](https://github.com/galipremsagar) +- Add requirements for rmm ([#739](https://github.com/rapidsai/rmm/pull/739)) [@galipremsagar](https://github.com/galipremsagar) +- device_uvector can be used within thrust::optional ([#734](https://github.com/rapidsai/rmm/pull/734)) [@robertmaynard](https://github.com/robertmaynard) +- arena_memory_resource optimization: disable tracking allocated blocks by default ([#732](https://github.com/rapidsai/rmm/pull/732)) [@rongou](https://github.com/rongou) +- Remove CMAKE_CURRENT_BINARY_DIR path in rmm's target_include_directories ([#731](https://github.com/rapidsai/rmm/pull/731)) [@trxcllnt](https://github.com/trxcllnt) +- set CMAKE_CUDA_ARCHITECTURES to OFF instead of undefined ([#729](https://github.com/rapidsai/rmm/pull/729)) [@trxcllnt](https://github.com/trxcllnt) +- Avoid potential race conditions in device_scalar/device_uvector setters ([#725](https://github.com/rapidsai/rmm/pull/725)) [@harrism](https://github.com/harrism) +- Update Changelog Link ([#723](https://github.com/rapidsai/rmm/pull/723)) [@ajschmidt8](https://github.com/ajschmidt8) +- Prepare Changelog for Automation ([#717](https://github.com/rapidsai/rmm/pull/717)) [@ajschmidt8](https://github.com/ajschmidt8) +- Update 0.18 changelog entry ([#716](https://github.com/rapidsai/rmm/pull/716)) [@ajschmidt8](https://github.com/ajschmidt8) +- Simplify cmake cuda architectures handling ([#709](https://github.com/rapidsai/rmm/pull/709)) [@robertmaynard](https://github.com/robertmaynard) +- Build only `compute` for the newest arch in CMAKE_CUDA_ARCHITECTURES ([#706](https://github.com/rapidsai/rmm/pull/706)) [@robertmaynard](https://github.com/robertmaynard) +- ENH Build with Ninja & Pass ccache variables to conda recipe ([#705](https://github.com/rapidsai/rmm/pull/705)) [@dillon-cullinan](https://github.com/dillon-cullinan) +- pool_memory_resource optimization: disable tracking allocated blocks by default ([#702](https://github.com/rapidsai/rmm/pull/702)) [@harrism](https://github.com/harrism) +- Allow the build directory of rmm to be used for `find_package(rmm)` ([#698](https://github.com/rapidsai/rmm/pull/698)) [@robertmaynard](https://github.com/robertmaynard) +- Adds a linear accessor to RMM cuda stream pool ([#696](https://github.com/rapidsai/rmm/pull/696)) [@afender](https://github.com/afender) +- Fix merge conflicts for #692 ([#694](https://github.com/rapidsai/rmm/pull/694)) [@ajschmidt8](https://github.com/ajschmidt8) +- Fix merge conflicts for #692 ([#693](https://github.com/rapidsai/rmm/pull/693)) [@ajschmidt8](https://github.com/ajschmidt8) +- Remove C++ Wrappers in `memory_resource_adaptors.hpp` Needed by Cython ([#662](https://github.com/rapidsai/rmm/pull/662)) [@mdemoret-nv](https://github.com/mdemoret-nv) +- Improve Cython Lifetime Management by Adding References in `DeviceBuffer` ([#661](https://github.com/rapidsai/rmm/pull/661)) [@mdemoret-nv](https://github.com/mdemoret-nv) +- Add support for streams in CuPy allocator ([#654](https://github.com/rapidsai/rmm/pull/654)) [@pentschev](https://github.com/pentschev) # RMM 0.18.0 (24 Feb 2021)