Known Issues

GELS produces the wrong result with sequential version

Problem Description:

 

We have a number of customers report regarding incorrect behavior of GELS in Intel® MKL 11.0 if sequential library is linked.

Customer’s quote:”When I compile and run this program with Intel® Fortran Composer XE 2013(**), the output is totally wrong. Compiling and running with Intel FORTRAN Composer XE 2011 (*) gives the correct results.”

  • Intermediate
  • Intel® C++ Composer XE
  • Intel® Fortran Composer XE
  • Intel® Math Kernel Library
  • Statically linking MKL library and IPP library in same project produce the link errors

    Problem Description:     Statically linking MKL and IPP in the same project produce the link errors like the following:

    1>ipps_l.lib(pscopyg9as_g9.obj) : error LNK2005: _g9_ownsSet_32s_G9 already defined in mkl_core.lib(pscopyg9as_20120907.obj)

    1>ipps_l.lib(pscopyg9as_g9.obj) : error LNK2005: _g9_ownsSet_16u_G9 already defined in mkl_core.lib(pscopyg9as_20120907.obj)

    1>ipps_l.lib(pscopyg9as_g9.obj) : error LNK2005: _g9_ownsSet_8u_G9 already defined in mkl_core.lib(pscopyg9as_20120907.obj)

  • Developers
  • C/C++
  • Fortran
  • Intermediate
  • Intel® Integrated Performance Primitives
  • Intel® Math Kernel Library
  • What Caused the issues on Intel Android* Jelly Bean System Image Source Unzip and Decompression When Using WinRAR or Winzip on Windows?

    You are a developer using a Windows system ( Windows 8, 7 or even older) to develop your Android* app. You want to use Intel Android* system image source to compile your own version of Android* for either emulator or VirtualBox.

    Building NCAR Graphics* with the Intel® compilers


    Introduction

    This guide is intended to help Intel® C++ Compiler and Intel® Fortran Compiler customers build the NCAR Graphics* library. NCAR Graphics is an open source library for drawing maps, graphs, and plots. More information about NCAR Graphics can be found at http://ngwww.ucar.edu


    Version information

    NCAR Graphics 6.1.2
    Intel® C++ Compiler for Linux* 2013 SP1 Update 2
    Intel® Fortran Compiler for Linux* 2013 SP1 Update 2

  • Linux*
  • C/C++
  • Fortran
  • Intel® C++ Compiler
  • Intel® Fortran Compiler
  • Parallel Computing
  • MKL VSL example for grouppooledcovariance (vsldgrouppooledcovariance.f) failed with PGI threading layer.

    Problem Description:

    Grouppooledcovariance example crashs with segmentation fault with PGI threading layer on all platforms due to error in PGI support of OpenMP order clause.

    We can see this example “vsldgrouppooledcovariance.f” into the following directory MKL_ROOT>\examples\vslf\source\". This example shows how to calculates of group/poopled covariances.

     The issue number is DPD2002622493

     Root Cause:

  • Developers
  • Students
  • Linux*
  • Unix*
  • Fortran
  • Intermediate
  • Intel® Math Kernel Library
  • OpenMP*
  • PGI
  • VSL
  • Covariance
  • vslssseditpooledcovariance
  • vsldsseditpooledcovariance
  • MKL VSL
  • Don’t use multiple runs with stack collection for event-based sampling

    Problem:

    VTune™ Amplifier XE 2013 supports new feature called “Call Stack Collection” for hardware PMU event-based sampling. However if the user set option “Allow multiple runs” in vtune’s project properties and enable “Call Stack Collection” option, to run some predefined analysis such as snb-general-exploration, snb-memory-access, etc. It will cause segmentation fail. Example command line:

  • Developers
  • Linux*
  • Beginner
  • Intel® VTune™ Amplifier XE
  • VTune Amplifier XE Call Stack Multiple Runs
  • Development Tools
  • Workaround for interoperability issue with Symantec* Endpoint Client Protection software older than 12.1.2015.2015 for Intel(R) Advisor XE 2013 survey and Intel(R) VTune(TM) Amplifier XE 2013 hotspot analyses

    Problem

    Intel(R) Advisor XE 2013 Survey and Intel(R) VTune(TM) Amplifier XE 2013 hotspot analyses fails when running on a system with Symantec* Endpoint Client Protection software versions older than 12.1.2015.2015 (For ex: 12.1.1000.157).

    Symptoms

    As soon as analysis is started, Application Under Test crashes experiences a crash or other sorts of failures. Our tools may report one or more of the following error messages:

  • Microsoft Windows* (XP, Vista, 7)
  • Intel® Advisor XE
  • Intel® VTune™ Amplifier XE
  • MPI sets LD_LIBRARY_PATH incorrectly on Xeon Phi™ under LSF*

    Symptom

    Shared libraries are not found when running a rank on a Xeon Phi™ coprocessor.

    error while loading shared libraries: lib*.so: cannot open shared object file: No such file or directory

    Cause

    When launching a job running in either native or symmetric mode, ranks launched on Xeon Phi" coprocessors under the LSF* job scheduler will not correctly set LD_LIBRARY_PATH.  Instead, the default LD_LIBRARY_PATH for the coprocessor will be used.  If shared libraries are located in other locations, they will not be found.

  • Developers
  • Linux*
  • Server
  • Intermediate
  • Intel® MPI Library
  • Intel® Cluster Studio XE
  • Message Passing Interface
  • LSF
  • HYDRA
  • LD_LIBRARY_PATH
  • Intel® Many Integrated Core Architecture
  • Subscribe to Known Issues