LAPACK  3.10.0
LAPACK: Linear Algebra PACKage
cdrvhp.f
Go to the documentation of this file.
1 *> \brief \b CDRVHP
2 *
3 * =========== DOCUMENTATION ===========
4 *
5 * Online html documentation available at
6 * http://www.netlib.org/lapack/explore-html/
7 *
8 * Definition:
9 * ===========
10 *
11 * SUBROUTINE CDRVHP( DOTYPE, NN, NVAL, NRHS, THRESH, TSTERR, NMAX,
12 * A, AFAC, AINV, B, X, XACT, WORK, RWORK, IWORK,
13 * NOUT )
14 *
15 * .. Scalar Arguments ..
16 * LOGICAL TSTERR
17 * INTEGER NMAX, NN, NOUT, NRHS
18 * REAL THRESH
19 * ..
20 * .. Array Arguments ..
21 * LOGICAL DOTYPE( * )
22 * INTEGER IWORK( * ), NVAL( * )
23 * REAL RWORK( * )
24 * COMPLEX A( * ), AFAC( * ), AINV( * ), B( * ),
25 * $ WORK( * ), X( * ), XACT( * )
26 * ..
27 *
28 *
29 *> \par Purpose:
30 * =============
31 *>
32 *> \verbatim
33 *>
34 *> CDRVHP tests the driver routines CHPSV and -SVX.
35 *> \endverbatim
36 *
37 * Arguments:
38 * ==========
39 *
40 *> \param[in] DOTYPE
41 *> \verbatim
42 *> DOTYPE is LOGICAL array, dimension (NTYPES)
43 *> The matrix types to be used for testing. Matrices of type j
44 *> (for 1 <= j <= NTYPES) are used for testing if DOTYPE(j) =
45 *> .TRUE.; if DOTYPE(j) = .FALSE., then type j is not used.
46 *> \endverbatim
47 *>
48 *> \param[in] NN
49 *> \verbatim
50 *> NN is INTEGER
51 *> The number of values of N contained in the vector NVAL.
52 *> \endverbatim
53 *>
54 *> \param[in] NVAL
55 *> \verbatim
56 *> NVAL is INTEGER array, dimension (NN)
57 *> The values of the matrix dimension N.
58 *> \endverbatim
59 *>
60 *> \param[in] NRHS
61 *> \verbatim
62 *> NRHS is INTEGER
63 *> The number of right hand side vectors to be generated for
64 *> each linear system.
65 *> \endverbatim
66 *>
67 *> \param[in] THRESH
68 *> \verbatim
69 *> THRESH is REAL
70 *> The threshold value for the test ratios. A result is
71 *> included in the output file if RESULT >= THRESH. To have
72 *> every test ratio printed, use THRESH = 0.
73 *> \endverbatim
74 *>
75 *> \param[in] TSTERR
76 *> \verbatim
77 *> TSTERR is LOGICAL
78 *> Flag that indicates whether error exits are to be tested.
79 *> \endverbatim
80 *>
81 *> \param[in] NMAX
82 *> \verbatim
83 *> NMAX is INTEGER
84 *> The maximum value permitted for N, used in dimensioning the
85 *> work arrays.
86 *> \endverbatim
87 *>
88 *> \param[out] A
89 *> \verbatim
90 *> A is COMPLEX array, dimension
91 *> (NMAX*(NMAX+1)/2)
92 *> \endverbatim
93 *>
94 *> \param[out] AFAC
95 *> \verbatim
96 *> AFAC is COMPLEX array, dimension
97 *> (NMAX*(NMAX+1)/2)
98 *> \endverbatim
99 *>
100 *> \param[out] AINV
101 *> \verbatim
102 *> AINV is COMPLEX array, dimension
103 *> (NMAX*(NMAX+1)/2)
104 *> \endverbatim
105 *>
106 *> \param[out] B
107 *> \verbatim
108 *> B is COMPLEX array, dimension (NMAX*NRHS)
109 *> \endverbatim
110 *>
111 *> \param[out] X
112 *> \verbatim
113 *> X is COMPLEX array, dimension (NMAX*NRHS)
114 *> \endverbatim
115 *>
116 *> \param[out] XACT
117 *> \verbatim
118 *> XACT is COMPLEX array, dimension (NMAX*NRHS)
119 *> \endverbatim
120 *>
121 *> \param[out] WORK
122 *> \verbatim
123 *> WORK is COMPLEX array, dimension
124 *> (NMAX*max(2,NRHS))
125 *> \endverbatim
126 *>
127 *> \param[out] RWORK
128 *> \verbatim
129 *> RWORK is REAL array, dimension (NMAX+2*NRHS)
130 *> \endverbatim
131 *>
132 *> \param[out] IWORK
133 *> \verbatim
134 *> IWORK is INTEGER array, dimension (NMAX)
135 *> \endverbatim
136 *>
137 *> \param[in] NOUT
138 *> \verbatim
139 *> NOUT is INTEGER
140 *> The unit number for output.
141 *> \endverbatim
142 *
143 * Authors:
144 * ========
145 *
146 *> \author Univ. of Tennessee
147 *> \author Univ. of California Berkeley
148 *> \author Univ. of Colorado Denver
149 *> \author NAG Ltd.
150 *
151 *> \ingroup complex_lin
152 *
153 * =====================================================================
154  SUBROUTINE cdrvhp( DOTYPE, NN, NVAL, NRHS, THRESH, TSTERR, NMAX,
155  $ A, AFAC, AINV, B, X, XACT, WORK, RWORK, IWORK,
156  $ NOUT )
157 *
158 * -- LAPACK test routine --
159 * -- LAPACK is a software package provided by Univ. of Tennessee, --
160 * -- Univ. of California Berkeley, Univ. of Colorado Denver and NAG Ltd..--
161 *
162 * .. Scalar Arguments ..
163  LOGICAL TSTERR
164  INTEGER NMAX, NN, NOUT, NRHS
165  REAL THRESH
166 * ..
167 * .. Array Arguments ..
168  LOGICAL DOTYPE( * )
169  INTEGER IWORK( * ), NVAL( * )
170  REAL RWORK( * )
171  COMPLEX A( * ), AFAC( * ), AINV( * ), B( * ),
172  $ work( * ), x( * ), xact( * )
173 * ..
174 *
175 * =====================================================================
176 *
177 * .. Parameters ..
178  REAL ONE, ZERO
179  PARAMETER ( ONE = 1.0e+0, zero = 0.0e+0 )
180  INTEGER NTYPES, NTESTS
181  parameter( ntypes = 10, ntests = 6 )
182  INTEGER NFACT
183  parameter( nfact = 2 )
184 * ..
185 * .. Local Scalars ..
186  LOGICAL ZEROT
187  CHARACTER DIST, FACT, PACKIT, TYPE, UPLO, XTYPE
188  CHARACTER*3 PATH
189  INTEGER I, I1, I2, IFACT, IMAT, IN, INFO, IOFF, IUPLO,
190  $ izero, j, k, k1, kl, ku, lda, mode, n, nb,
191  $ nbmin, nerrs, nfail, nimat, npp, nrun, nt
192  REAL AINVNM, ANORM, CNDNUM, RCOND, RCONDC
193 * ..
194 * .. Local Arrays ..
195  CHARACTER FACTS( NFACT )
196  INTEGER ISEED( 4 ), ISEEDY( 4 )
197  REAL RESULT( NTESTS )
198 * ..
199 * .. External Functions ..
200  REAL CLANHP, SGET06
201  EXTERNAL CLANHP, SGET06
202 * ..
203 * .. External Subroutines ..
204  EXTERNAL aladhd, alaerh, alasvm, ccopy, cerrvx, cget04,
207  $ cppt05, xlaenv
208 * ..
209 * .. Scalars in Common ..
210  LOGICAL LERR, OK
211  CHARACTER*32 SRNAMT
212  INTEGER INFOT, NUNIT
213 * ..
214 * .. Common blocks ..
215  COMMON / infoc / infot, nunit, ok, lerr
216  COMMON / srnamc / srnamt
217 * ..
218 * .. Intrinsic Functions ..
219  INTRINSIC cmplx, max, min
220 * ..
221 * .. Data statements ..
222  DATA iseedy / 1988, 1989, 1990, 1991 /
223  DATA facts / 'F', 'N' /
224 * ..
225 * .. Executable Statements ..
226 *
227 * Initialize constants and the random number seed.
228 *
229  path( 1: 1 ) = 'C'
230  path( 2: 3 ) = 'HP'
231  nrun = 0
232  nfail = 0
233  nerrs = 0
234  DO 10 i = 1, 4
235  iseed( i ) = iseedy( i )
236  10 CONTINUE
237 *
238 * Test the error exits
239 *
240  IF( tsterr )
241  $ CALL cerrvx( path, nout )
242  infot = 0
243 *
244 * Set the block size and minimum block size for testing.
245 *
246  nb = 1
247  nbmin = 2
248  CALL xlaenv( 1, nb )
249  CALL xlaenv( 2, nbmin )
250 *
251 * Do for each value of N in NVAL
252 *
253  DO 180 in = 1, nn
254  n = nval( in )
255  lda = max( n, 1 )
256  npp = n*( n+1 ) / 2
257  xtype = 'N'
258  nimat = ntypes
259  IF( n.LE.0 )
260  $ nimat = 1
261 *
262  DO 170 imat = 1, nimat
263 *
264 * Do the tests only if DOTYPE( IMAT ) is true.
265 *
266  IF( .NOT.dotype( imat ) )
267  $ GO TO 170
268 *
269 * Skip types 3, 4, 5, or 6 if the matrix size is too small.
270 *
271  zerot = imat.GE.3 .AND. imat.LE.6
272  IF( zerot .AND. n.LT.imat-2 )
273  $ GO TO 170
274 *
275 * Do first for UPLO = 'U', then for UPLO = 'L'
276 *
277  DO 160 iuplo = 1, 2
278  IF( iuplo.EQ.1 ) THEN
279  uplo = 'U'
280  packit = 'C'
281  ELSE
282  uplo = 'L'
283  packit = 'R'
284  END IF
285 *
286 * Set up parameters with CLATB4 and generate a test matrix
287 * with CLATMS.
288 *
289  CALL clatb4( path, imat, n, n, TYPE, kl, ku, anorm, mode,
290  $ cndnum, dist )
291 *
292  srnamt = 'CLATMS'
293  CALL clatms( n, n, dist, iseed, TYPE, rwork, mode,
294  $ cndnum, anorm, kl, ku, packit, a, lda, work,
295  $ info )
296 *
297 * Check error code from CLATMS.
298 *
299  IF( info.NE.0 ) THEN
300  CALL alaerh( path, 'CLATMS', info, 0, uplo, n, n, -1,
301  $ -1, -1, imat, nfail, nerrs, nout )
302  GO TO 160
303  END IF
304 *
305 * For types 3-6, zero one or more rows and columns of the
306 * matrix to test that INFO is returned correctly.
307 *
308  IF( zerot ) THEN
309  IF( imat.EQ.3 ) THEN
310  izero = 1
311  ELSE IF( imat.EQ.4 ) THEN
312  izero = n
313  ELSE
314  izero = n / 2 + 1
315  END IF
316 *
317  IF( imat.LT.6 ) THEN
318 *
319 * Set row and column IZERO to zero.
320 *
321  IF( iuplo.EQ.1 ) THEN
322  ioff = ( izero-1 )*izero / 2
323  DO 20 i = 1, izero - 1
324  a( ioff+i ) = zero
325  20 CONTINUE
326  ioff = ioff + izero
327  DO 30 i = izero, n
328  a( ioff ) = zero
329  ioff = ioff + i
330  30 CONTINUE
331  ELSE
332  ioff = izero
333  DO 40 i = 1, izero - 1
334  a( ioff ) = zero
335  ioff = ioff + n - i
336  40 CONTINUE
337  ioff = ioff - izero
338  DO 50 i = izero, n
339  a( ioff+i ) = zero
340  50 CONTINUE
341  END IF
342  ELSE
343  ioff = 0
344  IF( iuplo.EQ.1 ) THEN
345 *
346 * Set the first IZERO rows and columns to zero.
347 *
348  DO 70 j = 1, n
349  i2 = min( j, izero )
350  DO 60 i = 1, i2
351  a( ioff+i ) = zero
352  60 CONTINUE
353  ioff = ioff + j
354  70 CONTINUE
355  ELSE
356 *
357 * Set the last IZERO rows and columns to zero.
358 *
359  DO 90 j = 1, n
360  i1 = max( j, izero )
361  DO 80 i = i1, n
362  a( ioff+i ) = zero
363  80 CONTINUE
364  ioff = ioff + n - j
365  90 CONTINUE
366  END IF
367  END IF
368  ELSE
369  izero = 0
370  END IF
371 *
372 * Set the imaginary part of the diagonals.
373 *
374  IF( iuplo.EQ.1 ) THEN
375  CALL claipd( n, a, 2, 1 )
376  ELSE
377  CALL claipd( n, a, n, -1 )
378  END IF
379 *
380  DO 150 ifact = 1, nfact
381 *
382 * Do first for FACT = 'F', then for other values.
383 *
384  fact = facts( ifact )
385 *
386 * Compute the condition number for comparison with
387 * the value returned by CHPSVX.
388 *
389  IF( zerot ) THEN
390  IF( ifact.EQ.1 )
391  $ GO TO 150
392  rcondc = zero
393 *
394  ELSE IF( ifact.EQ.1 ) THEN
395 *
396 * Compute the 1-norm of A.
397 *
398  anorm = clanhp( '1', uplo, n, a, rwork )
399 *
400 * Factor the matrix A.
401 *
402  CALL ccopy( npp, a, 1, afac, 1 )
403  CALL chptrf( uplo, n, afac, iwork, info )
404 *
405 * Compute inv(A) and take its norm.
406 *
407  CALL ccopy( npp, afac, 1, ainv, 1 )
408  CALL chptri( uplo, n, ainv, iwork, work, info )
409  ainvnm = clanhp( '1', uplo, n, ainv, rwork )
410 *
411 * Compute the 1-norm condition number of A.
412 *
413  IF( anorm.LE.zero .OR. ainvnm.LE.zero ) THEN
414  rcondc = one
415  ELSE
416  rcondc = ( one / anorm ) / ainvnm
417  END IF
418  END IF
419 *
420 * Form an exact solution and set the right hand side.
421 *
422  srnamt = 'CLARHS'
423  CALL clarhs( path, xtype, uplo, ' ', n, n, kl, ku,
424  $ nrhs, a, lda, xact, lda, b, lda, iseed,
425  $ info )
426  xtype = 'C'
427 *
428 * --- Test CHPSV ---
429 *
430  IF( ifact.EQ.2 ) THEN
431  CALL ccopy( npp, a, 1, afac, 1 )
432  CALL clacpy( 'Full', n, nrhs, b, lda, x, lda )
433 *
434 * Factor the matrix and solve the system using CHPSV.
435 *
436  srnamt = 'CHPSV '
437  CALL chpsv( uplo, n, nrhs, afac, iwork, x, lda,
438  $ info )
439 *
440 * Adjust the expected value of INFO to account for
441 * pivoting.
442 *
443  k = izero
444  IF( k.GT.0 ) THEN
445  100 CONTINUE
446  IF( iwork( k ).LT.0 ) THEN
447  IF( iwork( k ).NE.-k ) THEN
448  k = -iwork( k )
449  GO TO 100
450  END IF
451  ELSE IF( iwork( k ).NE.k ) THEN
452  k = iwork( k )
453  GO TO 100
454  END IF
455  END IF
456 *
457 * Check error code from CHPSV .
458 *
459  IF( info.NE.k ) THEN
460  CALL alaerh( path, 'CHPSV ', info, k, uplo, n,
461  $ n, -1, -1, nrhs, imat, nfail,
462  $ nerrs, nout )
463  GO TO 120
464  ELSE IF( info.NE.0 ) THEN
465  GO TO 120
466  END IF
467 *
468 * Reconstruct matrix from factors and compute
469 * residual.
470 *
471  CALL chpt01( uplo, n, a, afac, iwork, ainv, lda,
472  $ rwork, result( 1 ) )
473 *
474 * Compute residual of the computed solution.
475 *
476  CALL clacpy( 'Full', n, nrhs, b, lda, work, lda )
477  CALL cppt02( uplo, n, nrhs, a, x, lda, work, lda,
478  $ rwork, result( 2 ) )
479 *
480 * Check solution from generated exact solution.
481 *
482  CALL cget04( n, nrhs, x, lda, xact, lda, rcondc,
483  $ result( 3 ) )
484  nt = 3
485 *
486 * Print information about the tests that did not pass
487 * the threshold.
488 *
489  DO 110 k = 1, nt
490  IF( result( k ).GE.thresh ) THEN
491  IF( nfail.EQ.0 .AND. nerrs.EQ.0 )
492  $ CALL aladhd( nout, path )
493  WRITE( nout, fmt = 9999 )'CHPSV ', uplo, n,
494  $ imat, k, result( k )
495  nfail = nfail + 1
496  END IF
497  110 CONTINUE
498  nrun = nrun + nt
499  120 CONTINUE
500  END IF
501 *
502 * --- Test CHPSVX ---
503 *
504  IF( ifact.EQ.2 .AND. npp.GT.0 )
505  $ CALL claset( 'Full', npp, 1, cmplx( zero ),
506  $ cmplx( zero ), afac, npp )
507  CALL claset( 'Full', n, nrhs, cmplx( zero ),
508  $ cmplx( zero ), x, lda )
509 *
510 * Solve the system and compute the condition number and
511 * error bounds using CHPSVX.
512 *
513  srnamt = 'CHPSVX'
514  CALL chpsvx( fact, uplo, n, nrhs, a, afac, iwork, b,
515  $ lda, x, lda, rcond, rwork,
516  $ rwork( nrhs+1 ), work, rwork( 2*nrhs+1 ),
517  $ info )
518 *
519 * Adjust the expected value of INFO to account for
520 * pivoting.
521 *
522  k = izero
523  IF( k.GT.0 ) THEN
524  130 CONTINUE
525  IF( iwork( k ).LT.0 ) THEN
526  IF( iwork( k ).NE.-k ) THEN
527  k = -iwork( k )
528  GO TO 130
529  END IF
530  ELSE IF( iwork( k ).NE.k ) THEN
531  k = iwork( k )
532  GO TO 130
533  END IF
534  END IF
535 *
536 * Check the error code from CHPSVX.
537 *
538  IF( info.NE.k ) THEN
539  CALL alaerh( path, 'CHPSVX', info, k, fact // uplo,
540  $ n, n, -1, -1, nrhs, imat, nfail,
541  $ nerrs, nout )
542  GO TO 150
543  END IF
544 *
545  IF( info.EQ.0 ) THEN
546  IF( ifact.GE.2 ) THEN
547 *
548 * Reconstruct matrix from factors and compute
549 * residual.
550 *
551  CALL chpt01( uplo, n, a, afac, iwork, ainv, lda,
552  $ rwork( 2*nrhs+1 ), result( 1 ) )
553  k1 = 1
554  ELSE
555  k1 = 2
556  END IF
557 *
558 * Compute residual of the computed solution.
559 *
560  CALL clacpy( 'Full', n, nrhs, b, lda, work, lda )
561  CALL cppt02( uplo, n, nrhs, a, x, lda, work, lda,
562  $ rwork( 2*nrhs+1 ), result( 2 ) )
563 *
564 * Check solution from generated exact solution.
565 *
566  CALL cget04( n, nrhs, x, lda, xact, lda, rcondc,
567  $ result( 3 ) )
568 *
569 * Check the error bounds from iterative refinement.
570 *
571  CALL cppt05( uplo, n, nrhs, a, b, lda, x, lda,
572  $ xact, lda, rwork, rwork( nrhs+1 ),
573  $ result( 4 ) )
574  ELSE
575  k1 = 6
576  END IF
577 *
578 * Compare RCOND from CHPSVX with the computed value
579 * in RCONDC.
580 *
581  result( 6 ) = sget06( rcond, rcondc )
582 *
583 * Print information about the tests that did not pass
584 * the threshold.
585 *
586  DO 140 k = k1, 6
587  IF( result( k ).GE.thresh ) THEN
588  IF( nfail.EQ.0 .AND. nerrs.EQ.0 )
589  $ CALL aladhd( nout, path )
590  WRITE( nout, fmt = 9998 )'CHPSVX', fact, uplo,
591  $ n, imat, k, result( k )
592  nfail = nfail + 1
593  END IF
594  140 CONTINUE
595  nrun = nrun + 7 - k1
596 *
597  150 CONTINUE
598 *
599  160 CONTINUE
600  170 CONTINUE
601  180 CONTINUE
602 *
603 * Print a summary of the results.
604 *
605  CALL alasvm( path, nout, nfail, nrun, nerrs )
606 *
607  9999 FORMAT( 1x, a, ', UPLO=''', a1, ''', N =', i5, ', type ', i2,
608  $ ', test ', i2, ', ratio =', g12.5 )
609  9998 FORMAT( 1x, a, ', FACT=''', a1, ''', UPLO=''', a1, ''', N =', i5,
610  $ ', type ', i2, ', test ', i2, ', ratio =', g12.5 )
611  RETURN
612 *
613 * End of CDRVHP
614 *
615  END
subroutine alasvm(TYPE, NOUT, NFAIL, NRUN, NERRS)
ALASVM
Definition: alasvm.f:73
subroutine xlaenv(ISPEC, NVALUE)
XLAENV
Definition: xlaenv.f:81
subroutine aladhd(IOUNIT, PATH)
ALADHD
Definition: aladhd.f:90
subroutine alaerh(PATH, SUBNAM, INFO, INFOE, OPTS, M, N, KL, KU, N5, IMAT, NFAIL, NERRS, NOUT)
ALAERH
Definition: alaerh.f:147
subroutine ccopy(N, CX, INCX, CY, INCY)
CCOPY
Definition: ccopy.f:81
subroutine clarhs(PATH, XTYPE, UPLO, TRANS, M, N, KL, KU, NRHS, A, LDA, X, LDX, B, LDB, ISEED, INFO)
CLARHS
Definition: clarhs.f:208
subroutine cppt02(UPLO, N, NRHS, A, X, LDX, B, LDB, RWORK, RESID)
CPPT02
Definition: cppt02.f:123
subroutine cdrvhp(DOTYPE, NN, NVAL, NRHS, THRESH, TSTERR, NMAX, A, AFAC, AINV, B, X, XACT, WORK, RWORK, IWORK, NOUT)
CDRVHP
Definition: cdrvhp.f:157
subroutine clatb4(PATH, IMAT, M, N, TYPE, KL, KU, ANORM, MODE, CNDNUM, DIST)
CLATB4
Definition: clatb4.f:121
subroutine cget04(N, NRHS, X, LDX, XACT, LDXACT, RCOND, RESID)
CGET04
Definition: cget04.f:102
subroutine cppt05(UPLO, N, NRHS, AP, B, LDB, X, LDX, XACT, LDXACT, FERR, BERR, RESLTS)
CPPT05
Definition: cppt05.f:157
subroutine claipd(N, A, INDA, VINDA)
CLAIPD
Definition: claipd.f:83
subroutine chpt01(UPLO, N, A, AFAC, IPIV, C, LDC, RWORK, RESID)
CHPT01
Definition: chpt01.f:113
subroutine cerrvx(PATH, NUNIT)
CERRVX
Definition: cerrvx.f:55
subroutine clatms(M, N, DIST, ISEED, SYM, D, MODE, COND, DMAX, KL, KU, PACK, A, LDA, WORK, INFO)
CLATMS
Definition: clatms.f:332
subroutine claset(UPLO, M, N, ALPHA, BETA, A, LDA)
CLASET initializes the off-diagonal elements and the diagonal elements of a matrix to given values.
Definition: claset.f:106
subroutine clacpy(UPLO, M, N, A, LDA, B, LDB)
CLACPY copies all or part of one two-dimensional array to another.
Definition: clacpy.f:103
subroutine chptrf(UPLO, N, AP, IPIV, INFO)
CHPTRF
Definition: chptrf.f:159
subroutine chptri(UPLO, N, AP, IPIV, WORK, INFO)
CHPTRI
Definition: chptri.f:109
subroutine chpsvx(FACT, UPLO, N, NRHS, AP, AFP, IPIV, B, LDB, X, LDX, RCOND, FERR, BERR, WORK, RWORK, INFO)
CHPSVX computes the solution to system of linear equations A * X = B for OTHER matrices
Definition: chpsvx.f:277
subroutine chpsv(UPLO, N, NRHS, AP, IPIV, B, LDB, INFO)
CHPSV computes the solution to system of linear equations A * X = B for OTHER matrices
Definition: chpsv.f:162