Parallel Performance
Argument |
Comment |
| -qopenmp | Causes multi-threaded code to be generated when OpenMP directives are present. For Fortran only, makes local arrays automatic and may require an increased stack size. |
| -parallel | The auto-parallelizer detects simply structured loops that may be safely executed in parallel and automatically generates multi-threaded code for these loops. |
| -par-threshold[n] | Sets a threshold for the auto-parallelization of loops based on the likelihood of a performance benefit. n=0 to 100, default 100. 0: Parallelize loops regardless of computation work volume. 100: Parallelize loops only if a performance benefit is highly likely. Must be used in conjunction with -parallel. |
| -par-affinity=name | Specifies thread-processor affinity for OpenMP or auto-parallelized applications. Typical values of name are none (default), scatter and compact. |
| -guide[=n] | Guided Auto-Parallelization. Causes the compiler to suggest ways to help loops to vectorize or auto-parallelize, without producing any objects or executables. n is an optional value from 1 to 4 specifying increasing levels of guidance to be provided, level 4 being the most advanced and aggressive. If n is omitted, the default is 4. |
| -q[no-]opt-matmul | This option enables [disables] a compiler-generated Matrix Multiply (matmul) library call by identifying matrix multiplication loop nests, if any, and replacing them with a matmul library call for improved performance. This option is enabled by default if options -O3 and -parallel are specified. This option has no effect unless option -O2 or higher is set. |
| -cilk-serialize | This option includes a header file, cilk_stubs.h, that causes the compiler to ignore all Intel® Cilk Plus threading keywords, resulting in a serial executable. (C/C++ only). |
| -coarray=shared | Enables the coarray feature of Fortran 2008 on shared memory systems. |
| -mkl=name | Requests that the Intel® Math Kernel Library (Intel® MKL) be linked. Off by default. Possible values of name are: parallel: Links the threaded part of Intel® MKL (default) sequential: Links the non-threaded part of Intel® MKL cluster: Links the cluster and sequential parts of Intel® MKL |
