Name

umf_lusolve — solve a linear sparse system given the LU factors

Calling Sequence

[x] = umf_lusolve(LU_ptr, b [, st, A])

Parameters

LU_ptr

a pointer to umf lu factors (L,U,p,q,R)

b

a real or complex column vector or a matrix (multiple rhs)

st

(optional) a string "Ax=b" (default) or "Ax'=b" (to be written "Ax''=b" in scilab langage: a quote in a string must be doubled !)

A

(optional) the sparse square matrix corresponding to the LU factors (LU_ptr must be got with LU_ptr = umf_lufact(A))

x

a column vector or a matrix in case of multiple rhs ( x(:,i) is solution of A x(:,i) = b(:,i) or A'x(:,i) = b(:,i) )

Description

This function must be used in conjonction with umf_lufact which computes the LU factors of a sparse matrix. The optional st argument lets us choose between the solving of Ax=b (general case) or of A'x=b (sometimes useful). If you give the 4th argument then iterative refinement will be also proceceed (as in umfpack) to give a better numerical solution.

Examples

see the example section of umf_lufact

See Also

umfpack , umf_lufact , umf_luget , umf_ludel , umf_luinfo

Authors

umfpack by Timothy A. Davis (see umf_license)
scilab interface by Bruno Pincon with contributions from Antonio Frasson