csc_ilu csc_ilu0 csc_ilu_solve csc_ssor_solve csr_ilu csr_ilu0 csr_ilu_solve csr_ssor_solve csx_ds csx_ds_solve csx_ssor

csx_ssor() void csx_ssor ( int  n, double  omega, const double  val[], const int  ptr[], const int  ind[], int  base, double  d[], int  iwork[], int *  info  ) Initialize symmetric successive over-relaxation (SSOR) preconditioner (CSC/CSR) Purpose This routine initializes the symmetric successive over-relaxation (SSOR) preconditioner for the coefficient matrix A of the sparse linear equations. The preconditioner matrix is defined as follows. M(ω) = (1/(2 - ω))(D/ω + L)(D/ω)^(-1)(D/ω + U) where L, D and U are the lower triangular part, the diagonal part and the upper triangular part of the coefficient matrix A, respectively (If A is a symmetric matrix, U = L^T). The output d[] is used by csc_ssor_solve(), csr_ssor_solve(), ssc_ssor_solve() or ssr_ssor_solve(). Parameters [in] n Dimension of matrix A. (n >= 0) (If n = 0, returns without computation) [in] omega The relaxation parameter ω. (0 < ω < 2) [in] val[] Array val[lval] (lval >= nnz) Values of non-zero elements of matrix A. (nnz is number of non-zero elements) [in] ptr[] Array ptr[lptr] (lptr >= n + 1) Column pointers (CSC) / row pointers (CSR) of matrix A. [in] ind[] Array ind[lind] (lind >= nnz) Row indices (CSC) / column indices (CSR) of matrix A. (nnz is number of non-zero elements) [in] base Indexing of ptr[] and ind[]. = 0: Zero-based (C style) indexing: Starting index is 0. = 1: One-based (Fortran style) indexing: Starting index is 1. [out] d[] Array d[ld] (ld >= n) Diagonal elements D/ω. [out] iwork[] Array iwork[liwork] (liwork >= n) Work array. [out] info = 0: Successful exit. = i < 0: The (-i)-th argument is invalid. = j > 0: j diagonal elements are zero (Returns 1/ω in d[] for such elements).