Returns the number of Intel Xeon Phi coprocessors on the system when called on the host CPU.
Floating point computations are not exact on almost all modern architectures. This lack of precision is particularly problematic in parallel operations. Since floating point computations are inexact, algorithms are classified according to whether they are repeatable and to what degree they guarantee coherence.
Modifies a pointer to the memory representing a coordinate of the data stored in matrix format.