LAPACK  3.10.0
LAPACK: Linear Algebra PACKage

◆ cspt03()

subroutine cspt03 ( character  UPLO,
integer  N,
complex, dimension( * )  A,
complex, dimension( * )  AINV,
complex, dimension( ldw, * )  WORK,
integer  LDW,
real, dimension( * )  RWORK,
real  RCOND,
real  RESID 
)

CSPT03

Purpose:
 CSPT03 computes the residual for a complex symmetric packed matrix
 times its inverse:
    norm( I - A*AINV ) / ( N * norm(A) * norm(AINV) * EPS ),
 where EPS is the machine epsilon.
Parameters
[in]UPLO
          UPLO is CHARACTER*1
          Specifies whether the upper or lower triangular part of the
          complex symmetric matrix A is stored:
          = 'U':  Upper triangular
          = 'L':  Lower triangular
[in]N
          N is INTEGER
          The number of rows and columns of the matrix A.  N >= 0.
[in]A
          A is COMPLEX array, dimension (N*(N+1)/2)
          The original complex symmetric matrix A, stored as a packed
          triangular matrix.
[in]AINV
          AINV is COMPLEX array, dimension (N*(N+1)/2)
          The (symmetric) inverse of the matrix A, stored as a packed
          triangular matrix.
[out]WORK
          WORK is COMPLEX array, dimension (LDW,N)
[in]LDW
          LDW is INTEGER
          The leading dimension of the array WORK.  LDW >= max(1,N).
[out]RWORK
          RWORK is REAL array, dimension (N)
[out]RCOND
          RCOND is REAL
          The reciprocal of the condition number of A, computed as
          ( 1/norm(A) ) / norm(AINV).
[out]RESID
          RESID is REAL
          norm(I - A*AINV) / ( N * norm(A) * norm(AINV) * EPS )
Author
Univ. of Tennessee
Univ. of California Berkeley
Univ. of Colorado Denver
NAG Ltd.

Definition at line 108 of file cspt03.f.

110 *
111 * -- LAPACK test routine --
112 * -- LAPACK is a software package provided by Univ. of Tennessee, --
113 * -- Univ. of California Berkeley, Univ. of Colorado Denver and NAG Ltd..--
114 *
115 * .. Scalar Arguments ..
116  CHARACTER UPLO
117  INTEGER LDW, N
118  REAL RCOND, RESID
119 * ..
120 * .. Array Arguments ..
121  REAL RWORK( * )
122  COMPLEX A( * ), AINV( * ), WORK( LDW, * )
123 * ..
124 *
125 * =====================================================================
126 *
127 * .. Parameters ..
128  REAL ZERO, ONE
129  parameter( zero = 0.0e+0, one = 1.0e+0 )
130 * ..
131 * .. Local Scalars ..
132  INTEGER I, ICOL, J, JCOL, K, KCOL, NALL
133  REAL AINVNM, ANORM, EPS
134  COMPLEX T
135 * ..
136 * .. External Functions ..
137  LOGICAL LSAME
138  REAL CLANGE, CLANSP, SLAMCH
139  COMPLEX CDOTU
140  EXTERNAL lsame, clange, clansp, slamch, cdotu
141 * ..
142 * .. Intrinsic Functions ..
143  INTRINSIC real
144 * ..
145 * .. Executable Statements ..
146 *
147 * Quick exit if N = 0.
148 *
149  IF( n.LE.0 ) THEN
150  rcond = one
151  resid = zero
152  RETURN
153  END IF
154 *
155 * Exit with RESID = 1/EPS if ANORM = 0 or AINVNM = 0.
156 *
157  eps = slamch( 'Epsilon' )
158  anorm = clansp( '1', uplo, n, a, rwork )
159  ainvnm = clansp( '1', uplo, n, ainv, rwork )
160  IF( anorm.LE.zero .OR. ainvnm.LE.zero ) THEN
161  rcond = zero
162  resid = one / eps
163  RETURN
164  END IF
165  rcond = ( one/anorm ) / ainvnm
166 *
167 * Case where both A and AINV are upper triangular:
168 * Each element of - A * AINV is computed by taking the dot product
169 * of a row of A with a column of AINV.
170 *
171  IF( lsame( uplo, 'U' ) ) THEN
172  DO 70 i = 1, n
173  icol = ( ( i-1 )*i ) / 2 + 1
174 *
175 * Code when J <= I
176 *
177  DO 30 j = 1, i
178  jcol = ( ( j-1 )*j ) / 2 + 1
179  t = cdotu( j, a( icol ), 1, ainv( jcol ), 1 )
180  jcol = jcol + 2*j - 1
181  kcol = icol - 1
182  DO 10 k = j + 1, i
183  t = t + a( kcol+k )*ainv( jcol )
184  jcol = jcol + k
185  10 CONTINUE
186  kcol = kcol + 2*i
187  DO 20 k = i + 1, n
188  t = t + a( kcol )*ainv( jcol )
189  kcol = kcol + k
190  jcol = jcol + k
191  20 CONTINUE
192  work( i, j ) = -t
193  30 CONTINUE
194 *
195 * Code when J > I
196 *
197  DO 60 j = i + 1, n
198  jcol = ( ( j-1 )*j ) / 2 + 1
199  t = cdotu( i, a( icol ), 1, ainv( jcol ), 1 )
200  jcol = jcol - 1
201  kcol = icol + 2*i - 1
202  DO 40 k = i + 1, j
203  t = t + a( kcol )*ainv( jcol+k )
204  kcol = kcol + k
205  40 CONTINUE
206  jcol = jcol + 2*j
207  DO 50 k = j + 1, n
208  t = t + a( kcol )*ainv( jcol )
209  kcol = kcol + k
210  jcol = jcol + k
211  50 CONTINUE
212  work( i, j ) = -t
213  60 CONTINUE
214  70 CONTINUE
215  ELSE
216 *
217 * Case where both A and AINV are lower triangular
218 *
219  nall = ( n*( n+1 ) ) / 2
220  DO 140 i = 1, n
221 *
222 * Code when J <= I
223 *
224  icol = nall - ( ( n-i+1 )*( n-i+2 ) ) / 2 + 1
225  DO 100 j = 1, i
226  jcol = nall - ( ( n-j )*( n-j+1 ) ) / 2 - ( n-i )
227  t = cdotu( n-i+1, a( icol ), 1, ainv( jcol ), 1 )
228  kcol = i
229  jcol = j
230  DO 80 k = 1, j - 1
231  t = t + a( kcol )*ainv( jcol )
232  jcol = jcol + n - k
233  kcol = kcol + n - k
234  80 CONTINUE
235  jcol = jcol - j
236  DO 90 k = j, i - 1
237  t = t + a( kcol )*ainv( jcol+k )
238  kcol = kcol + n - k
239  90 CONTINUE
240  work( i, j ) = -t
241  100 CONTINUE
242 *
243 * Code when J > I
244 *
245  icol = nall - ( ( n-i )*( n-i+1 ) ) / 2
246  DO 130 j = i + 1, n
247  jcol = nall - ( ( n-j+1 )*( n-j+2 ) ) / 2 + 1
248  t = cdotu( n-j+1, a( icol-n+j ), 1, ainv( jcol ), 1 )
249  kcol = i
250  jcol = j
251  DO 110 k = 1, i - 1
252  t = t + a( kcol )*ainv( jcol )
253  jcol = jcol + n - k
254  kcol = kcol + n - k
255  110 CONTINUE
256  kcol = kcol - i
257  DO 120 k = i, j - 1
258  t = t + a( kcol+k )*ainv( jcol )
259  jcol = jcol + n - k
260  120 CONTINUE
261  work( i, j ) = -t
262  130 CONTINUE
263  140 CONTINUE
264  END IF
265 *
266 * Add the identity matrix to WORK .
267 *
268  DO 150 i = 1, n
269  work( i, i ) = work( i, i ) + one
270  150 CONTINUE
271 *
272 * Compute norm(I - A*AINV) / (N * norm(A) * norm(AINV) * EPS)
273 *
274  resid = clange( '1', n, n, work, ldw, rwork )
275 *
276  resid = ( ( resid*rcond )/eps ) / real( n )
277 *
278  RETURN
279 *
280 * End of CSPT03
281 *
logical function lsame(CA, CB)
LSAME
Definition: lsame.f:53
complex function cdotu(N, CX, INCX, CY, INCY)
CDOTU
Definition: cdotu.f:83
real function clange(NORM, M, N, A, LDA, WORK)
CLANGE returns the value of the 1-norm, Frobenius norm, infinity-norm, or the largest absolute value ...
Definition: clange.f:115
real function clansp(NORM, UPLO, N, AP, WORK)
CLANSP returns the value of the 1-norm, or the Frobenius norm, or the infinity norm,...
Definition: clansp.f:115
real function slamch(CMACH)
SLAMCH
Definition: slamch.f:68
Here is the caller graph for this function: