The release of the PoCL 1.3 (Portable Computing Language OpenCL) project is available, which develops an implementation of the OpenCL standard that is independent of graphics accelerator manufacturers and allows you to use different backends to run OpenCL cores on different types of graphics and central processors. The project code is distributed under the MIT license. Work is supported on platforms X86_64, MIPS32, ARM v7, AMD HSA APU and various specialized TTA processors (Transport Triggered Architecture) with VLIW architecture.
The OpenCL core compiler implementation is based on LLVM, and Clang is used as a front-end for OpenCL C. To ensure proper portability and performance, the OpenCL kernel compiler can generate combined functions that can use various hardware resources to parallelize code execution, such as VLIW, superscalar, SIMD, SIMT, multicore, and multithreading. Support for ICD drivers
(Installable Client Driver). There are backends to support work through the CPU, ASIP (TCE / TTA), GPU based on the HSA architecture and GPU NVIDIA (CUDA).
The new version adds support for LLVM/Clang 8.0. Provided support for ICD (Installable Client Driver) on the macOS platform. Implemented the ability to build pocl without backend drivers for the CPU. For HSA (Heterogeneous System Architecture) initial support for compiling ISA over HSA runtime is provided. The code base was cleaned up, including the removal of the Vecmathlib library and the end of support for older versions of LLVM (less than 6.0).
Source: opennet.ru