1SGERFS(1)                LAPACK routine (version 3.2)                SGERFS(1)
2
3
4

NAME

6       SGERFS - improves the computed solution to a system of linear equations
7       and provides error bounds and backward error estimates for the solution
8

SYNOPSIS

10       SUBROUTINE SGERFS( TRANS, N, NRHS, A, LDA, AF, LDAF, IPIV, B,  LDB,  X,
11                          LDX, FERR, BERR, WORK, IWORK, INFO )
12
13           CHARACTER      TRANS
14
15           INTEGER        INFO, LDA, LDAF, LDB, LDX, N, NRHS
16
17           INTEGER        IPIV( * ), IWORK( * )
18
19           REAL           A(  LDA, * ), AF( LDAF, * ), B( LDB, * ), BERR( * ),
20                          FERR( * ), WORK( * ), X( LDX, * )
21

PURPOSE

23       SGERFS improves the computed solution to a system of  linear  equations
24       and  provides  error  bounds and backward error estimates for the solu‐
25       tion.
26

ARGUMENTS

28       TRANS   (input) CHARACTER*1
29               Specifies the form of the system of equations:
30               = 'N':  A * X = B     (No transpose)
31               = 'T':  A**T * X = B  (Transpose)
32               = 'C':  A**H * X = B  (Conjugate transpose = Transpose)
33
34       N       (input) INTEGER
35               The order of the matrix A.  N >= 0.
36
37       NRHS    (input) INTEGER
38               The number of right hand sides, i.e., the number of columns  of
39               the matrices B and X.  NRHS >= 0.
40
41       A       (input) REAL array, dimension (LDA,N)
42               The original N-by-N matrix A.
43
44       LDA     (input) INTEGER
45               The leading dimension of the array A.  LDA >= max(1,N).
46
47       AF      (input) REAL array, dimension (LDAF,N)
48               The  factors  L  and U from the factorization A = P*L*U as com‐
49               puted by SGETRF.
50
51       LDAF    (input) INTEGER
52               The leading dimension of the array AF.  LDAF >= max(1,N).
53
54       IPIV    (input) INTEGER array, dimension (N)
55               The pivot indices from SGETRF; for 1<=i<=N, row i of the matrix
56               was interchanged with row IPIV(i).
57
58       B       (input) REAL array, dimension (LDB,NRHS)
59               The right hand side matrix B.
60
61       LDB     (input) INTEGER
62               The leading dimension of the array B.  LDB >= max(1,N).
63
64       X       (input/output) REAL array, dimension (LDX,NRHS)
65               On  entry,  the  solution  matrix X, as computed by SGETRS.  On
66               exit, the improved solution matrix X.
67
68       LDX     (input) INTEGER
69               The leading dimension of the array X.  LDX >= max(1,N).
70
71       FERR    (output) REAL array, dimension (NRHS)
72               The estimated forward error bound for each solution vector X(j)
73               (the  j-th  column  of the solution matrix X).  If XTRUE is the
74               true solution corresponding to X(j), FERR(j)  is  an  estimated
75               upper bound for the magnitude of the largest element in (X(j) -
76               XTRUE) divided by the magnitude of the largest element in X(j).
77               The  estimate  is as reliable as the estimate for RCOND, and is
78               almost always a slight overestimate of the true error.
79
80       BERR    (output) REAL array, dimension (NRHS)
81               The componentwise relative backward error of each solution vec‐
82               tor  X(j) (i.e., the smallest relative change in any element of
83               A or B that makes X(j) an exact solution).
84
85       WORK    (workspace) REAL array, dimension (3*N)
86
87       IWORK   (workspace) INTEGER array, dimension (N)
88
89       INFO    (output) INTEGER
90               = 0:  successful exit
91               < 0:  if INFO = -i, the i-th argument had an illegal value
92

PARAMETERS

94       ITMAX is the maximum number of steps of iterative refinement.
95
96
97
98 LAPACK routine (version 3.2)    November 2008                       SGERFS(1)