For data parallel versions of Fortran, such as CM Fortran and Cray Fortran for the T3D, arrays are provided within the language which assume a given data distribution across processors. In these cases, arrays such as A, U and B passed to PCG are assumed to be such distributed arrays.
A great deal of flexibility is available for the rank, size and distribution of such arrays. However, the arrays must satisfy certain alignment conditions in order for high performance to be achieved. Furthermore, arrays passed to user routines <mysuba> and <mysubq> are guaranteed to match the arrays U or B in rank, size and distribution.
Since the Fortran 90 rules for array sequence and storage association are typically violated by data parallel Fortrans for parallel machines, certain modifications are sometimes necessary. For example, for the reverse communication interface, the parameters IVA, IVQL and IVQR for most versions are integer pointers into FWK but for data parallel versions must be actual arrays of appropriate rank, size and distribution.