opencl_src Directory Reference

Files
file	sigmaodd.cl [code]
	OpenCL implementation to check the sigma_odd problem, mostly like the other implementations but without the advantage of the shortcut in the factorization. If your GPU supports not too old OpenCL version, use the specific function ctz into divide_until_odd() instead the manual computation. Some tips used: "optimizing specifically for Processor Graphics OpenCL* device, ensure all conditionals are evaluated outside of code branches" https://software.intel.com/sites/landingpage/opencl/optimization-guide/Notes_on_Branching_Loops.htm.