December 2017: pocl v1.0 released
- Support for LLVM/Clang 5.0 and 4.0.
- Support for NVIDIA GPUs via a new CUDA backend (currently experimental)
- Most of the OpenCL 1.2 standard conformance tests pass with the CPU
backend. There are some caveats though - see the documentation.
Please note that there's now an official pocl maintenance policy in place. This text describes the policy and how you can get your favourite project that uses OpenCL to remain regression free in the future pocl releases.
We are looking for active maintainers to look after the ARM(64) and MacOS ports. If you are interested in helping to keep pocl working well on these platforms, please let us know!
Most of the code that landed to the pocl code base during this release cycle was produced for the needs of research funded by various sources. The Customized Parallel Computing research group of Tampere University of Technology (Finland) likes to thank the Academy of Finland (funding decision 297548), Finnish Funding Agency for Technology and Innovation (funding decision 40142/14, FiDiPro-StreamPro), ARTEMIS joint undertaking under grant agreement no 621439 (ALMARVI) and HSA Foundation.