MXM
  C/OpenMP version.

  Matrix multiplication tests.

  The number of processors available:
  OMP_GET_NUM_PROCS () = 2

  The number of threads is 2

R8_MXM matrix multiplication timing.
  A(LxN) = B(LxM) * C(MxN).
  L = 500
  M = 500
  N = 500
  Floating point OPS roughly 250000000
  Elapsed time dT = 2.635515
  Rate = MegaOPS/dT = 94.858122

MXM:
  Normal end of execution.