1       SUBROUTINE CTRT02( UPLO, TRANS, DIAG, N, NRHS, A, LDA, X, LDX, B,
  2      $                   LDB, WORK, RWORK, RESID )
  3 *
  4 *  -- LAPACK test routine (version 3.1) --
  5 *     Univ. of Tennessee, Univ. of California Berkeley and NAG Ltd..
  6 *     November 2006
  7 *
  8 *     .. Scalar Arguments ..
  9       CHARACTER          DIAG, TRANS, UPLO
 10       INTEGER            LDA, LDB, LDX, N, NRHS
 11       REAL               RESID
 12 *     ..
 13 *     .. Array Arguments ..
 14       REAL               RWORK( * )
 15       COMPLEX            A( LDA, * ), B( LDB, * ), WORK( * ),
 16      $                   X( LDX, * )
 17 *     ..
 18 *
 19 *  Purpose
 20 *  =======
 21 *
 22 *  CTRT02 computes the residual for the computed solution to a
 23 *  triangular system of linear equations  A*x = b,  A**T *x = b,
 24 *  or A**H *x = b.  Here A is a triangular matrix, A**T is the transpose
 25 *  of A, A**H is the conjugate transpose of A, and x and b are N by NRHS
 26 *  matrices.  The test ratio is the maximum over the number of right
 27 *  hand sides of
 28 *     norm(b - op(A)*x) / ( norm(op(A)) * norm(x) * EPS ),
 29 *  where op(A) denotes A, A**T, or A**H, and EPS is the machine epsilon.
 30 *
 31 *  Arguments
 32 *  =========
 33 *
 34 *  UPLO    (input) CHARACTER*1
 35 *          Specifies whether the matrix A is upper or lower triangular.
 36 *          = 'U':  Upper triangular
 37 *          = 'L':  Lower triangular
 38 *
 39 *  TRANS   (input) CHARACTER*1
 40 *          Specifies the operation applied to A.
 41 *          = 'N':  A *x = b     (No transpose)
 42 *          = 'T':  A**T *x = b  (Transpose)
 43 *          = 'C':  A**H *x = b  (Conjugate transpose)
 44 *
 45 *  DIAG    (input) CHARACTER*1
 46 *          Specifies whether or not the matrix A is unit triangular.
 47 *          = 'N':  Non-unit triangular
 48 *          = 'U':  Unit triangular
 49 *
 50 *  N       (input) INTEGER
 51 *          The order of the matrix A.  N >= 0.
 52 *
 53 *  NRHS    (input) INTEGER
 54 *          The number of right hand sides, i.e., the number of columns
 55 *          of the matrices X and B.  NRHS >= 0.
 56 *
 57 *  A       (input) COMPLEX array, dimension (LDA,N)
 58 *          The triangular matrix A.  If UPLO = 'U', the leading n by n
 59 *          upper triangular part of the array A contains the upper
 60 *          triangular matrix, and the strictly lower triangular part of
 61 *          A is not referenced.  If UPLO = 'L', the leading n by n lower
 62 *          triangular part of the array A contains the lower triangular
 63 *          matrix, and the strictly upper triangular part of A is not
 64 *          referenced.  If DIAG = 'U', the diagonal elements of A are
 65 *          also not referenced and are assumed to be 1.
 66 *
 67 *  LDA     (input) INTEGER
 68 *          The leading dimension of the array A.  LDA >= max(1,N).
 69 *
 70 *  X       (input) COMPLEX array, dimension (LDX,NRHS)
 71 *          The computed solution vectors for the system of linear
 72 *          equations.
 73 *
 74 *  LDX     (input) INTEGER
 75 *          The leading dimension of the array X.  LDX >= max(1,N).
 76 *
 77 *  B       (input) COMPLEX array, dimension (LDB,NRHS)
 78 *          The right hand side vectors for the system of linear
 79 *          equations.
 80 *
 81 *  LDB     (input) INTEGER
 82 *          The leading dimension of the array B.  LDB >= max(1,N).
 83 *
 84 *  WORK    (workspace) COMPLEX array, dimension (N)
 85 *
 86 *  RWORK   (workspace) REAL array, dimension (N)
 87 *
 88 *  RESID   (output) REAL
 89 *          The maximum over the number of right hand sides of
 90 *          norm(op(A)*x - b) / ( norm(op(A)) * norm(x) * EPS ).
 91 *
 92 *  =====================================================================
 93 *
 94 *     .. Parameters ..
 95       REAL               ZERO, ONE
 96       PARAMETER          ( ZERO = 0.0E+0, ONE = 1.0E+0 )
 97 *     ..
 98 *     .. Local Scalars ..
 99       INTEGER            J
100       REAL               ANORM, BNORM, EPS, XNORM
101 *     ..
102 *     .. External Functions ..
103       LOGICAL            LSAME
104       REAL               CLANTR, SCASUM, SLAMCH
105       EXTERNAL           LSAME, CLANTR, SCASUM, SLAMCH
106 *     ..
107 *     .. External Subroutines ..
108       EXTERNAL           CAXPY, CCOPY, CTRMV
109 *     ..
110 *     .. Intrinsic Functions ..
111       INTRINSIC          CMPLXMAX
112 *     ..
113 *     .. Executable Statements ..
114 *
115 *     Quick exit if N = 0 or NRHS = 0
116 *
117       IF( N.LE.0 .OR. NRHS.LE.0 ) THEN
118          RESID = ZERO
119          RETURN
120       END IF
121 *
122 *     Compute the 1-norm of A or A**H.
123 *
124       IF( LSAME( TRANS, 'N' ) ) THEN
125          ANORM = CLANTR( '1', UPLO, DIAG, N, N, A, LDA, RWORK )
126       ELSE
127          ANORM = CLANTR( 'I', UPLO, DIAG, N, N, A, LDA, RWORK )
128       END IF
129 *
130 *     Exit with RESID = 1/EPS if ANORM = 0.
131 *
132       EPS = SLAMCH( 'Epsilon' )
133       IF( ANORM.LE.ZERO ) THEN
134          RESID = ONE / EPS
135          RETURN
136       END IF
137 *
138 *     Compute the maximum over the number of right hand sides of
139 *        norm(op(A)*x - b) / ( norm(op(A)) * norm(x) * EPS )
140 *
141       RESID = ZERO
142       DO 10 J = 1, NRHS
143          CALL CCOPY( N, X( 1, J ), 1, WORK, 1 )
144          CALL CTRMV( UPLO, TRANS, DIAG, N, A, LDA, WORK, 1 )
145          CALL CAXPY( N, CMPLX-ONE ), B( 1, J ), 1, WORK, 1 )
146          BNORM = SCASUM( N, WORK, 1 )
147          XNORM = SCASUM( N, X( 1, J ), 1 )
148          IF( XNORM.LE.ZERO ) THEN
149             RESID = ONE / EPS
150          ELSE
151             RESID = MAX( RESID, ( ( BNORM / ANORM ) / XNORM ) / EPS )
152          END IF
153    10 CONTINUE
154 *
155       RETURN
156 *
157 *     End of CTRT02
158 *
159       END