Intel Software Adrenaline

Programming CodeBook

Issue link: http://read.uberflip.com/i/217587

Contents of this Issue

Navigation

Page 4 of 16

Compilation 5 COMPILATION Programming and optimizing for the Intel® Xeon Phi™ Coprocessor is a simple extension of programming and optimizing for the host processor. Intel® C, C++, and Fortran compilers support the same optimization features for the Intel® Xeon Phi™ coprocessor as for the host processor, including: • OpenMP* 4.0 for both native and offload models • Intel® Cilk™ Plus array notation • SIMD optimization • Vectorization • Expanded C++11 support • Expanded Fortran 2003 and 2008 support • Improved MPI performance and scalability • Loop optimizations, optimized math libraries, IPO Example of Intel® Cilk™ Plus Array Notation Loop was not vectorized: existence of vector dependence. LOOP WAS VECTORIZED. Example of Intel® Cilk™ Plus Pragma Loop was not vectorized: existence of vector dependence. SIMD LOOP WAS VECTORIZED. For more information regarding performance and optimization choices in Intel® software products, visit: http://software.intel.com/en-us/articles/optimization-notice Privacy | Terms of Use | Learn More

Articles in this issue

Links on this page

view archives of Intel Software Adrenaline - Programming CodeBook