TEST FAIL compilation MySQLCluster on Odroid Xu4 (ARM Ubuntu 16.04.02)/home/myLinuxUser/Download/mysql-cluster-gpl-7.5.6/CMakeFiles/CMakeError.log

CMakeError.log

root@vpn:/home/myLinuxUser/Download/mysql-cluster-gpl-7.5.6# cat /home/myLinuxUser/Download/mysql-cluster-gpl-7.5.6/CMakeFiles/CMakeError.log
Performing C SOURCE FILE Test HAVE_C_FLOATING_POINT_FUSED_MADD failed with the following output:
Change Dir: /home/myLinuxUser/Download/mysql-cluster-gpl-7.5.6/CMakeFiles/CMakeTmp

Run Build Command:"/usr/bin/make" "cmTC_b1685/fast"
/usr/bin/make -f CMakeFiles/cmTC_b1685.dir/build.make CMakeFiles/cmTC_b1685.dir/build
make[1]: Entering directory '/home/myLinuxUser/Download/mysql-cluster-gpl-7.5.6/CMakeFiles/CMakeTmp'
Building C object CMakeFiles/cmTC_b1685.dir/src.c.o
/usr/bin/cc    -DHAVE_C_FLOATING_POINT_FUSED_MADD  -O3   -o CMakeFiles/cmTC_b1685.dir/src.c.o   -c /home/myLinuxUser/Download/mysql-cluster-gpl-7.5.6/CMakeFiles/CMakeTmp/src.c
Linking C executable cmTC_b1685
/usr/bin/cmake -E cmake_link_script CMakeFiles/cmTC_b1685.dir/link.txt --verbose=1
/usr/bin/cc   -DHAVE_C_FLOATING_POINT_FUSED_MADD  -O3    CMakeFiles/cmTC_b1685.dir/src.c.o  -o cmTC_b1685 -rdynamic 
make[1]: Leaving directory '/home/myLinuxUser/Download/mysql-cluster-gpl-7.5.6/CMakeFiles/CMakeTmp'

Return value: 1
Source file was:

  int main (int argc, char **argv)
  {
    double n[21] = { 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
                     0, 0, 0, 1, 1, 1, 1, 1, 1, 1,1 };
    double m= 0, s= 0;
    int i;
    for(i= 0; i < 21; i++)
    {
      double m_kminusone= m;
      m= m_kminusone + (n[i] - m_kminusone) / (double) (i + 2);
      s= s + (n[i] - m_kminusone) * (n[i] - m);
    }
    /*
      s should now be either 5e 74 d1 45 17 5d 14 40 or
      40 14 5d 17 45 d1 74 5e, depending on endianness. If the floating point
      operations are optimized with fused multiply-add instructions, the least
      significant byte is 5d instead of 5e.
    */
    return (*(unsigned char*)(&s) == 0x5e ||
            *((unsigned char*)(&s) + 7) == 0x5e);
  }
Performing C++ SOURCE FILE Test HAVE_CXX_FLOATING_POINT_FUSED_MADD failed with the following output:
Change Dir: /home/myLinuxUser/Download/mysql-cluster-gpl-7.5.6/CMakeFiles/CMakeTmp

Run Build Command:"/usr/bin/make" "cmTC_c7c90/fast"
/usr/bin/make -f CMakeFiles/cmTC_c7c90.dir/build.make CMakeFiles/cmTC_c7c90.dir/build
make[1]: Entering directory '/home/myLinuxUser/Download/mysql-cluster-gpl-7.5.6/CMakeFiles/CMakeTmp'
Building CXX object CMakeFiles/cmTC_c7c90.dir/src.cxx.o
/usr/bin/c++     -DHAVE_CXX_FLOATING_POINT_FUSED_MADD  -O3   -o CMakeFiles/cmTC_c7c90.dir/src.cxx.o -c /home/myLinuxUser/Download/mysql-cluster-gpl-7.5.6/CMakeFiles/CMakeTmp/src.cxx
Linking CXX executable cmTC_c7c90
/usr/bin/cmake -E cmake_link_script CMakeFiles/cmTC_c7c90.dir/link.txt --verbose=1
/usr/bin/c++    -DHAVE_CXX_FLOATING_POINT_FUSED_MADD  -O3    CMakeFiles/cmTC_c7c90.dir/src.cxx.o  -o cmTC_c7c90 -rdynamic 
make[1]: Leaving directory '/home/myLinuxUser/Download/mysql-cluster-gpl-7.5.6/CMakeFiles/CMakeTmp'

Return value: 1
Source file was:

  int main (int argc, char **argv)
  {
    double n[21] = { 0, 0, 0, 0, 0, 0, 0, 0, 0, 0,
                     0, 0, 0, 1, 1, 1, 1, 1, 1, 1,1 };
    double m= 0, s= 0;
    int i;
    for(i= 0; i < 21; i++)
    {
      double m_kminusone= m;
      m= m_kminusone + (n[i] - m_kminusone) / (double) (i + 2);
      s= s + (n[i] - m_kminusone) * (n[i] - m);
    }
    /*
      s should now be either 5e 74 d1 45 17 5d 14 40 or
      40 14 5d 17 45 d1 74 5e, depending on endianness. If the floating point
      operations are optimized with fused multiply-add instructions, the least
      significant byte is 5d instead of 5e.
    */
    return (*(unsigned char*)(&s) == 0x5e ||
            *((unsigned char*)(&s) + 7) == 0x5e);
  }