Article

internal threshold was exceeded

In very large, complex functions, loops preceded by OpenMP directives may not be threaded. Compiler may emit the warning remark: "An internal threshold was exceeded: loops may not be vectorized or parallelized. Try to reduce routine size."
Authored by Martyn Corden (Intel) Last updated on 12/26/2018 - 13:11
Article

Building application with -openmp but no OpenMP pragmas causes segmentation fault

Compiling an application with the -openmp option that does not contain OpenMP pragmas causes a segmentation fault due to stack size overrun.
Authored by Elizabeth S (Intel) Last updated on 05/25/2018 - 15:30
Article

Intel® Professional Edition Compilers 11.1 Fixes List

This page lists fixes that have been made in Intel Professional Edition Compilers 11.1.
Authored by Elizabeth S (Intel) Last updated on 05/25/2018 - 15:30
Article

Threading Fortran Applications for Parallel Performance on Multi-Core Systems

Advice and background information is given on typical issues that may arise when threading an application using the Intel Fortran Compiler and other software tools, whether using OpenMP, automatic parallelization or threaded libraries.
Authored by Martyn Corden (Intel) Last updated on 12/12/2018 - 18:00
Article

Performance hints for WRF on Intel® architecture

This note explains how to configure the Weather Research & Forecasting (WRF) run-time environment in order to achieve best performance and scalability on Intel® architecture with Intel® software tools.
Authored by Roman Dubtsov (Intel) Last updated on 06/14/2017 - 13:03
Article

Using Tasking in OpenMP* with the Intel® Fortran Compiler

The multi-core performance of a legacy Fortran benchmark unsuited to data parallelism is enhanced by threading using the TASK construct of OpenMP and the Intel Fortran Compiler. The necessary source code changes are explained in detail.
Authored by Martyn Corden (Intel) Last updated on 07/04/2019 - 10:00
Article

Using KMP_AFFINITY to create OpenMP* thread mapping to OS proc IDs

This article explains how to create your own OpenMP* thread mapping to OS proc IDs with Intel® C/C++ and Fortran compiler, by setting environment variable KMP_AFFINITY. It also provides some thread affinity demo results on when use different settings.
Authored by Last updated on 05/25/2018 - 15:30
Article

Don't Use the OpenMP* THREADPRIVATE 'Compatibility' Option when Everything is Compiled by Intel

The Intel C++ and Fortran compilers for Windows* and Linux* provide 'legacy' and 'compatibility' implementations of the OpenMP THREADPRIVATE directive. The 'compatibility' option should not be used when everything is compiled by Intel compilers.
Authored by Kenneth Craft (Intel) Last updated on 07/08/2019 - 15:12
Article

Intel® C++ and Fortran Compilers now support the OpenMP* 3.1 Specification

The newest releases of the Intel® C++ and Fortran Compilers support new features in the OpenMP* 3.1 Specification
Authored by Kenneth Craft (Intel) Last updated on 05/25/2018 - 15:48
Article

Bogus OpenMP KMP_AFFINITY warnings on non-Intel processor hosts

An application compiled with -Qopenmp (Windows*)/-openmp (Linux*) and run on a host not containing a genuine Intel® processor may generate bogus OMP: Warning #71 and Warning #72.
Authored by Kenneth Craft (Intel) Last updated on 05/25/2018 - 15:48