POCL 1.3 Released For OpenCL On CPUs As The Portable Computing Language
Just days after the most recent POCL 1.3 release candidate, the Portable Computing Language 1.3 release is now officially available for an open-source project allowing OpenCL kernels to be executed on CPUs.
POCL 1.3 implements OpenCL 1.2 with some CL 2.0 functionality and is a portable OpenCL implementation for running on CPUs or there is also some offloading capabilities to NVIDIA GPUs via CUDA or AMD GPUs via HSA.
The POCL 1.3 release adds in LLVM/Clang 8.0 compiler support with LLVM doing much of the heavy-lifting for this portable OpenCL implementation, there is now installable client driver (ICD) support on macOS, HSA improvements, CUDA fixes, and many other fixes / code improvements throughout.
POCL 1.3 can be downloaded from PortableCL.org.
POCL 1.3 implements OpenCL 1.2 with some CL 2.0 functionality and is a portable OpenCL implementation for running on CPUs or there is also some offloading capabilities to NVIDIA GPUs via CUDA or AMD GPUs via HSA.
The POCL 1.3 release adds in LLVM/Clang 8.0 compiler support with LLVM doing much of the heavy-lifting for this portable OpenCL implementation, there is now installable client driver (ICD) support on macOS, HSA improvements, CUDA fixes, and many other fixes / code improvements throughout.
POCL 1.3 can be downloaded from PortableCL.org.
1 Comment