Intel Continues Improving Its SYCL Stack - Now Supports Ahead-Of-Time Compilation

The Khronos SYCL standard as a single-source C++-based programming model for OpenCL is one of the exciting elements for Intel's GPU compute plans with the forthcoming Xe graphics cards and fits into their oneAPI umbrella. They just released their SYCL Compiler and Runtimes 2019-12 release with numerous updates.

First up this new version of their SYCL compiler/run-time features opencl-aot as a new tool for offering ahead-of-time compilation of SYCL sources. The AoT compilation tool is geared for generating device-dependent OpenCL program binaries optimized out of SPIR-V. The optimized binaries are catered for Intel's architecture.

In addition to now supporting ahead-of-time compilation, there are new attributes supported for Intel FPGAs, support for splitting device code modules into multiple modules, support for FPGA device libraries within fat static libraries, kernels and program caches are now thread-safe, and a variety of other improvements and fixes.

More details on this updated Intel SYCL Compiler and Runtime stack via GitHub.