Benchmarking OpenCL On Intel Graphics With Beignet 1.3
Last week marked the release of Intel's Beignet 1.3, their open-source project implementing OpenCL acceleration atop modern CPUs with HD/Iris Graphics. Significant with Beignet 1.3 is that they've finally implemented OpenCL 2.0 support! OpenCL 2.0 is now available for Skylake hardware and newer. Beignet 1.3 also has other new features, runtime improvements, LLVM 3.9 support, new extensions, and much more. Thus time for some benchmarking of this new Beignet release.
Beignet 1.3 was already added to Intel's rolling-release Clear Linux distribution, so I used that for running this fresh round of Beignet OpenCL benchmarks. Beignet 1.3 is there along with Mesa 17.0-dev, LLVM 3.9.1, and the Linux 4.9 kernel. Thus quick and easy for getting a bleeding-edge look at the OpenCL potential for Intel CPUs with Beignet.
All of the OpenCL benchmarks for this article were run in a fully-automated and reproducible manner with the open-source Phoronix Test Suite benchmarking software. Several different systems were tested comprised of a Pentium G4400, Core i5 6500, Core i5 6600K, Core i3 7100, Core i5 7600K, Core i7 7700K. Xeon E3-1235L v5, and Xeon E3-1245 v5.