LAPACK 3.12.0
LAPACK: Linear Algebra PACKage
Loading...
Searching...
No Matches

◆ dckcsd()

subroutine dckcsd ( integer  nm,
integer, dimension( * )  mval,
integer, dimension( * )  pval,
integer, dimension( * )  qval,
integer  nmats,
integer, dimension( 4 )  iseed,
double precision  thresh,
integer  mmax,
double precision, dimension( * )  x,
double precision, dimension( * )  xf,
double precision, dimension( * )  u1,
double precision, dimension( * )  u2,
double precision, dimension( * )  v1t,
double precision, dimension( * )  v2t,
double precision, dimension( * )  theta,
integer, dimension( * )  iwork,
double precision, dimension( * )  work,
double precision, dimension( * )  rwork,
integer  nin,
integer  nout,
integer  info 
)

DCKCSD

Purpose:
 DCKCSD tests DORCSD:
        the CSD for an M-by-M orthogonal matrix X partitioned as
        [ X11 X12; X21 X22 ]. X11 is P-by-Q.
Parameters
[in]NM
          NM is INTEGER
          The number of values of M contained in the vector MVAL.
[in]MVAL
          MVAL is INTEGER array, dimension (NM)
          The values of the matrix row dimension M.
[in]PVAL
          PVAL is INTEGER array, dimension (NM)
          The values of the matrix row dimension P.
[in]QVAL
          QVAL is INTEGER array, dimension (NM)
          The values of the matrix column dimension Q.
[in]NMATS
          NMATS is INTEGER
          The number of matrix types to be tested for each combination
          of matrix dimensions.  If NMATS >= NTYPES (the maximum
          number of matrix types), then all the different types are
          generated for testing.  If NMATS < NTYPES, another input line
          is read to get the numbers of the matrix types to be used.
[in,out]ISEED
          ISEED is INTEGER array, dimension (4)
          On entry, the seed of the random number generator.  The array
          elements should be between 0 and 4095, otherwise they will be
          reduced mod 4096, and ISEED(4) must be odd.
          On exit, the next seed in the random number sequence after
          all the test matrices have been generated.
[in]THRESH
          THRESH is DOUBLE PRECISION
          The threshold value for the test ratios.  A result is
          included in the output file if RESULT >= THRESH.  To have
          every test ratio printed, use THRESH = 0.
[in]MMAX
          MMAX is INTEGER
          The maximum value permitted for M, used in dimensioning the
          work arrays.
[out]X
          X is DOUBLE PRECISION array, dimension (MMAX*MMAX)
[out]XF
          XF is DOUBLE PRECISION array, dimension (MMAX*MMAX)
[out]U1
          U1 is DOUBLE PRECISION array, dimension (MMAX*MMAX)
[out]U2
          U2 is DOUBLE PRECISION array, dimension (MMAX*MMAX)
[out]V1T
          V1T is DOUBLE PRECISION array, dimension (MMAX*MMAX)
[out]V2T
          V2T is DOUBLE PRECISION array, dimension (MMAX*MMAX)
[out]THETA
          THETA is DOUBLE PRECISION array, dimension (MMAX)
[out]IWORK
          IWORK is INTEGER array, dimension (MMAX)
[out]WORK
          WORK is DOUBLE PRECISION array
[out]RWORK
          RWORK is DOUBLE PRECISION array
[in]NIN
          NIN is INTEGER
          The unit number for input.
[in]NOUT
          NOUT is INTEGER
          The unit number for output.
[out]INFO
          INFO is INTEGER
          = 0 :  successful exit
          > 0 :  If DLAROR returns an error code, the absolute value
                 of it is returned.
Author
Univ. of Tennessee
Univ. of California Berkeley
Univ. of Colorado Denver
NAG Ltd.

Definition at line 181 of file dckcsd.f.

184*
185* -- LAPACK test routine --
186* -- LAPACK is a software package provided by Univ. of Tennessee, --
187* -- Univ. of California Berkeley, Univ. of Colorado Denver and NAG Ltd..--
188*
189* .. Scalar Arguments ..
190 INTEGER INFO, NIN, NM, NMATS, MMAX, NOUT
191 DOUBLE PRECISION THRESH
192* ..
193* .. Array Arguments ..
194 INTEGER ISEED( 4 ), IWORK( * ), MVAL( * ), PVAL( * ),
195 $ QVAL( * )
196 DOUBLE PRECISION RWORK( * ), THETA( * )
197 DOUBLE PRECISION U1( * ), U2( * ), V1T( * ), V2T( * ),
198 $ WORK( * ), X( * ), XF( * )
199* ..
200*
201* =====================================================================
202*
203* .. Parameters ..
204 INTEGER NTESTS
205 parameter( ntests = 15 )
206 INTEGER NTYPES
207 parameter( ntypes = 4 )
208 DOUBLE PRECISION GAPDIGIT, ONE, ORTH, TEN, ZERO
209 parameter( gapdigit = 18.0d0, one = 1.0d0,
210 $ orth = 1.0d-12,
211 $ ten = 10.0d0, zero = 0.0d0 )
212 DOUBLE PRECISION PIOVER2
213 parameter( piover2 = 1.57079632679489661923132169163975144210d0 )
214* ..
215* .. Local Scalars ..
216 LOGICAL FIRSTT
217 CHARACTER*3 PATH
218 INTEGER I, IINFO, IM, IMAT, J, LDU1, LDU2, LDV1T,
219 $ LDV2T, LDX, LWORK, M, NFAIL, NRUN, NT, P, Q, R
220* ..
221* .. Local Arrays ..
222 LOGICAL DOTYPE( NTYPES )
223 DOUBLE PRECISION RESULT( NTESTS )
224* ..
225* .. External Subroutines ..
226 EXTERNAL alahdg, alareq, alasum, dcsdts, dlacsg, dlaror,
227 $ dlaset, drot
228* ..
229* .. Intrinsic Functions ..
230 INTRINSIC abs, min
231* ..
232* .. External Functions ..
233 DOUBLE PRECISION DLARAN, DLARND
234 EXTERNAL dlaran, dlarnd
235* ..
236* .. Executable Statements ..
237*
238* Initialize constants and the random number seed.
239*
240 path( 1: 3 ) = 'CSD'
241 info = 0
242 nrun = 0
243 nfail = 0
244 firstt = .true.
245 CALL alareq( path, nmats, dotype, ntypes, nin, nout )
246 ldx = mmax
247 ldu1 = mmax
248 ldu2 = mmax
249 ldv1t = mmax
250 ldv2t = mmax
251 lwork = mmax*mmax
252*
253* Do for each value of M in MVAL.
254*
255 DO 30 im = 1, nm
256 m = mval( im )
257 p = pval( im )
258 q = qval( im )
259*
260 DO 20 imat = 1, ntypes
261*
262* Do the tests only if DOTYPE( IMAT ) is true.
263*
264 IF( .NOT.dotype( imat ) )
265 $ GO TO 20
266*
267* Generate X
268*
269 IF( imat.EQ.1 ) THEN
270 CALL dlaror( 'L', 'I', m, m, x, ldx, iseed, work, iinfo )
271 IF( m .NE. 0 .AND. iinfo .NE. 0 ) THEN
272 WRITE( nout, fmt = 9999 ) m, iinfo
273 info = abs( iinfo )
274 GO TO 20
275 END IF
276 ELSE IF( imat.EQ.2 ) THEN
277 r = min( p, m-p, q, m-q )
278 DO i = 1, r
279 theta(i) = piover2 * dlarnd( 1, iseed )
280 END DO
281 CALL dlacsg( m, p, q, theta, iseed, x, ldx, work )
282 DO i = 1, m
283 DO j = 1, m
284 x(i+(j-1)*ldx) = x(i+(j-1)*ldx) +
285 $ orth*dlarnd(2,iseed)
286 END DO
287 END DO
288 ELSE IF( imat.EQ.3 ) THEN
289 r = min( p, m-p, q, m-q )
290 DO i = 1, r+1
291 theta(i) = ten**(-dlarnd(1,iseed)*gapdigit)
292 END DO
293 DO i = 2, r+1
294 theta(i) = theta(i-1) + theta(i)
295 END DO
296 DO i = 1, r
297 theta(i) = piover2 * theta(i) / theta(r+1)
298 END DO
299 CALL dlacsg( m, p, q, theta, iseed, x, ldx, work )
300 ELSE
301 CALL dlaset( 'F', m, m, zero, one, x, ldx )
302 DO i = 1, m
303 j = int( dlaran( iseed ) * m ) + 1
304 IF( j .NE. i ) THEN
305 CALL drot( m, x(1+(i-1)*ldx), 1, x(1+(j-1)*ldx), 1,
306 $ zero, one )
307 END IF
308 END DO
309 END IF
310*
311 nt = 15
312*
313 CALL dcsdts( m, p, q, x, xf, ldx, u1, ldu1, u2, ldu2, v1t,
314 $ ldv1t, v2t, ldv2t, theta, iwork, work, lwork,
315 $ rwork, result )
316*
317* Print information about the tests that did not
318* pass the threshold.
319*
320 DO 10 i = 1, nt
321 IF( result( i ).GE.thresh ) THEN
322 IF( nfail.EQ.0 .AND. firstt ) THEN
323 firstt = .false.
324 CALL alahdg( nout, path )
325 END IF
326 WRITE( nout, fmt = 9998 )m, p, q, imat, i,
327 $ result( i )
328 nfail = nfail + 1
329 END IF
330 10 CONTINUE
331 nrun = nrun + nt
332 20 CONTINUE
333 30 CONTINUE
334*
335* Print a summary of the results.
336*
337 CALL alasum( path, nout, nfail, nrun, 0 )
338*
339 9999 FORMAT( ' DLAROR in DCKCSD: M = ', i5, ', INFO = ', i15 )
340 9998 FORMAT( ' M=', i4, ' P=', i4, ', Q=', i4, ', type ', i2,
341 $ ', test ', i2, ', ratio=', g13.6 )
342 RETURN
343*
344* End of DCKCSD
345*
subroutine alareq(path, nmats, dotype, ntypes, nin, nout)
ALAREQ
Definition alareq.f:90
subroutine alasum(type, nout, nfail, nrun, nerrs)
ALASUM
Definition alasum.f:73
subroutine alahdg(iounit, path)
ALAHDG
Definition alahdg.f:62
subroutine dlacsg(m, p, q, theta, iseed, x, ldx, work)
Definition dckcsd.f:351
subroutine dcsdts(m, p, q, x, xf, ldx, u1, ldu1, u2, ldu2, v1t, ldv1t, v2t, ldv2t, theta, iwork, work, lwork, rwork, result)
DCSDTS
Definition dcsdts.f:229
double precision function dlaran(iseed)
DLARAN
Definition dlaran.f:67
double precision function dlarnd(idist, iseed)
DLARND
Definition dlarnd.f:73
subroutine dlaror(side, init, m, n, a, lda, iseed, x, info)
DLAROR
Definition dlaror.f:146
subroutine dlaset(uplo, m, n, alpha, beta, a, lda)
DLASET initializes the off-diagonal elements and the diagonal elements of a matrix to given values.
Definition dlaset.f:110
subroutine drot(n, dx, incx, dy, incy, c, s)
DROT
Definition drot.f:92
Here is the call graph for this function:
Here is the caller graph for this function: