IV05125: INCORRECT RESULT WITH -QSMP=OMP

August 2011 PTF for XL C/C++ for AIX, V11.1
November 2011 PTF for XL C/C++ for AIX, V11.1
January 2012 PTF for XL C for AIX, V11.1
January 2012 PTF for XL C/C++ for AIX, V11.1
April 2012 PTF for XL C for AIX, V11.1
April 2012 PTF for XL C/C++ for AIX, V11.1
July 2012 PTF for XL C/C++ for AIX, V11.1
October 2012 PTF for XL C/C++ for AIX, V11.1
October 2012 PTF for XL C for AIX, V11.1
February 2013 PTF for XL C for AIX, V11.1
February 2013 PTF for XL C/C++ for AIX, V11.1
XL C/C++ for AIX Fix Pack 14 (May 2013 PTF) for 11.1
XL C for AIX Fix Pack 14 (May 2013 PTF) for 11.1
XL C/C++ for AIX Fix Pack 15 (August 2013 PTF) for 11.1
XL C for AIX Fix Pack 15 (August 2013 PTF) for 11.1
XL C/C++ for AIX Fix Pack 16 (November 2013 PTF) for 11.1
XL C for AIX Fix Pack 17 (February 2014 PTF) for 11.1
XL C/C++ for AIX Fix Pack 17 (February 2014 PTF) for 11.1
XL C/C++ for AIX Fix Pack 18 (May 2014 PTF) for 11.1
XL C/C++ for AIX Fix Pack 19 (August 2014 PTF) for 11.1
XL C/C++ for AIX Fix Pack 20 (November 2014 PTF) for 11.1
XL C for AIX Fix Pack 21 (April 2015 PTF) for 11.1
XL C/C++ for AIX Fix Pack 21 (April 2015 PTF) for 11.1
XL C/C++ for AIX Fix Pack 22 (September 2015 PTF) for 11.1
XL C for AIX Fix Pack 23 (March 2016 PTF) for 11.1
XL C/C++ for AIX Fix Pack 23 (March 2016 PTF) for 11.1
XL C for AIX Fix Pack 24 (September 2016 PTF) for 11.1
XL C/C++ for AIX Fix Pack 24 (September 2016 PTF) for 11.1
August 2011 PTF for XL C for AIX, V11.1
November 2011 PTF for XL C for AIX, V11.1
July 2012 PTF for XL C for AIX, V11.1
XL C for AIX Fix Pack 16 (November 2013 PTF) for 11.1
XL C for AIX Fix Pack 18 (May 2014 PTF) for 11.1
XL C for AIX Fix Pack 19 (August 2014 PTF) for 11.1
XL C for AIX Fix Pack 20 (November 2014 PTF) for 11.1
XL C for AIX Fix Pack 22 (September 2015 PTF) for 11.1

APAR status

• Closed as program error.

Error description

• The following test case gives out incorrect runtime results if
  compile with -qsmp=omp option.
  
  $cat simple_omp_bug.cpp
  #include <complex>
  #include <iostream>
  #include <vector>
  #ifdef _OPENMP
  # include "omp.h"
  #endif
  
  void Aggrevate
  ( std::complex<double> alpha, int height, int width,
    std::complex<double>* ABuf, const std::complex<double>* BBuf )
  {
  #ifdef _OPENMP
      #pragma omp parallel for
  #endif
      for( int j=0; j<width; ++j )
      {
          const std::complex<double>* BCol = &BBuf[j*height];
          std::complex<double>* ACol = &ABuf[j*height];
          for( int i=0; i<height; ++i )
              ACol[i] += alpha*BCol[i];
      }
  }
  
  int
  main( int argc, char* argv[] )
  {
      const int height = 100;
      const int width  = 100;
  
      std::vector< std::complex<double> > C(height*width,1.0);
      std::vector< std::complex<double> > D(height*width,0.7);
      std::complex<double> beta(0.3,0.);
  
      std::cout << "C[0] before: " << C[0] << "\n"
                << "D[0] before: " << D[0] << "\n"
                << "beta before: " << beta << "\n"
                << "height:      " << height << "\n"
                << "width:       " << width << "\n"
                << "C.size():    " << C.size() << "\n"
                << "D.size():    " << D.size() << std::endl;
  
      std::cout << "Started complex OpenMP section." << std::endl;
      Aggrevate( beta, height, width, &C[0], &D[0] );
      std::cout << "C[0] := C[0] + beta D[0] = " << C[0] <<
  std::endl;
  
      return 0;
  }
  
  ${bgxlC} -g -O0 simple_omp_bug.cpp -o simple_omp_bug_O0.x
  ${bgxlC_r} -qsmp=omp:noauto -g -O0 simple_omp_bug.cpp -o
  simple_omp_bug_omp_O0.x
  ${bgxlC} -g -O3 simple_omp_bug.cpp -o simple_omp_bug_O3.x
  ${bgxlC_r} -qsmp=omp:noauto -g -O3 simple_omp_bug.cpp -o
  simple_omp_bug_omp_O3.x
  ${bgxlC} -g -O4 simple_omp_bug.cpp -o simple_omp_bug_O4.x
  ${bgxlC_r} -qsmp=omp:noauto -g -O4 simple_omp_bug.cpp -o
  simple_omp_bug_omp_O4.x
  
  
  diff O0_omp.output O0.output
  9c9
  < C[0] := C[0] + beta D[0] = (1,4.10606e-313)
  ---
  > C[0] := C[0] + beta D[0] = (1.21,0)
  diff O3_omp.output
  O3.output
  9c9
  < C[0] := C[0] + beta D[0] = (1,4.10606e-313)
  ---
  > C[0] := C[0] + beta D[0] = (1.21,0)
  diff O4_omp.output
  O4.output
  9c9
  < C[0] := C[0] + beta D[0] = (1,0)
  ---
  > C[0] := C[0] + beta D[0] = (1.21,0)
  

Local fix

• N/A
  

Problem summary

• PROBLEM DESCRIPTION: Incorrect calculation of complex number
  with -qsmp=omp
  
  USERS AFFECTED: Complex numbers using -qsmp=omp to compile.
  

Problem conclusion

• In an outlined procedure the real and complex members of a
  complex number were not privatized into that procedure but the
  owner was. So the stores into them did not go into the location
  of the owner causing runtime failures when the owner was
  referenced.
  The problem is fixed. Apply recommended service.
  

Temporary fix

Comments

APAR Information

• APAR is sysrouted FROM one or more of the following:

  LI76052

• APAR is sysrouted TO one or more of the following:

Fix information

• Fixed component name

  XL C FOR AIX

• Fixed component ID

  5724X1200

Applicable component levels