LAPACK 3.11.0 LAPACK: Linear Algebra PACKage
Searching...
No Matches

◆ ddrvsy_rk()

 subroutine ddrvsy_rk ( logical, dimension( * ) DOTYPE, integer NN, integer, dimension( * ) NVAL, integer NRHS, double precision THRESH, logical TSTERR, integer NMAX, double precision, dimension( * ) A, double precision, dimension( * ) AFAC, double precision, dimension( * ) E, double precision, dimension( * ) AINV, double precision, dimension( * ) B, double precision, dimension( * ) X, double precision, dimension( * ) XACT, double precision, dimension( * ) WORK, double precision, dimension( * ) RWORK, integer, dimension( * ) IWORK, integer NOUT )

DDRVSY_RK

Purpose:
` DDRVSY_RK tests the driver routines DSYSV_RK.`
Parameters
 [in] DOTYPE ``` DOTYPE is LOGICAL array, dimension (NTYPES) The matrix types to be used for testing. Matrices of type j (for 1 <= j <= NTYPES) are used for testing if DOTYPE(j) = .TRUE.; if DOTYPE(j) = .FALSE., then type j is not used.``` [in] NN ``` NN is INTEGER The number of values of N contained in the vector NVAL.``` [in] NVAL ``` NVAL is INTEGER array, dimension (NN) The values of the matrix dimension N.``` [in] NRHS ``` NRHS is INTEGER The number of right hand side vectors to be generated for each linear system.``` [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] TSTERR ``` TSTERR is LOGICAL Flag that indicates whether error exits are to be tested.``` [in] NMAX ``` NMAX is INTEGER The maximum value permitted for N, used in dimensioning the work arrays.``` [out] A ` A is DOUBLE PRECISION array, dimension (NMAX*NMAX)` [out] AFAC ` AFAC is DOUBLE PRECISION array, dimension (NMAX*NMAX)` [out] E ` E is DOUBLE PRECISION array, dimension (NMAX)` [out] AINV ` AINV is DOUBLE PRECISION array, dimension (NMAX*NMAX)` [out] B ` B is DOUBLE PRECISION array, dimension (NMAX*NRHS)` [out] X ` X is DOUBLE PRECISION array, dimension (NMAX*NRHS)` [out] XACT ` XACT is DOUBLE PRECISION array, dimension (NMAX*NRHS)` [out] WORK ` WORK is DOUBLE PRECISION array, dimension (NMAX*max(2,NRHS))` [out] RWORK ` RWORK is DOUBLE PRECISION array, dimension (NMAX+2*NRHS)` [out] IWORK ` IWORK is INTEGER array, dimension (2*NMAX)` [in] NOUT ``` NOUT is INTEGER The unit number for output.```

Definition at line 153 of file ddrvsy_rk.f.

156*
157* -- LAPACK test routine --
158* -- LAPACK is a software package provided by Univ. of Tennessee, --
159* -- Univ. of California Berkeley, Univ. of Colorado Denver and NAG Ltd..--
160*
161* .. Scalar Arguments ..
162 LOGICAL TSTERR
163 INTEGER NMAX, NN, NOUT, NRHS
164 DOUBLE PRECISION THRESH
165* ..
166* .. Array Arguments ..
167 LOGICAL DOTYPE( * )
168 INTEGER IWORK( * ), NVAL( * )
169 DOUBLE PRECISION A( * ), AFAC( * ), AINV( * ), B( * ), E( * ),
170 \$ RWORK( * ), WORK( * ), X( * ), XACT( * )
171* ..
172*
173* =====================================================================
174*
175* .. Parameters ..
176 DOUBLE PRECISION ONE, ZERO
177 parameter( one = 1.0d+0, zero = 0.0d+0 )
178 INTEGER NTYPES, NTESTS
179 parameter( ntypes = 10, ntests = 3 )
180 INTEGER NFACT
181 parameter( nfact = 2 )
182* ..
183* .. Local Scalars ..
184 LOGICAL ZEROT
185 CHARACTER DIST, FACT, TYPE, UPLO, XTYPE
186 CHARACTER*3 PATH, MATPATH
187 INTEGER I, I1, I2, IFACT, IMAT, IN, INFO, IOFF, IUPLO,
188 \$ IZERO, J, K, KL, KU, LDA, LWORK, MODE, N,
189 \$ NB, NBMIN, NERRS, NFAIL, NIMAT, NRUN, NT
190 DOUBLE PRECISION AINVNM, ANORM, CNDNUM, RCONDC
191* ..
192* .. Local Arrays ..
193 CHARACTER FACTS( NFACT ), UPLOS( 2 )
194 INTEGER ISEED( 4 ), ISEEDY( 4 )
195 DOUBLE PRECISION RESULT( NTESTS )
196* ..
197* .. External Functions ..
198 DOUBLE PRECISION DLANSY
199 EXTERNAL dlansy
200* ..
201* .. External Subroutines ..
202 EXTERNAL aladhd, alaerh, alasvm, derrvx, dget04, dlacpy,
205* ..
206* .. Scalars in Common ..
207 LOGICAL LERR, OK
208 CHARACTER*32 SRNAMT
209 INTEGER INFOT, NUNIT
210* ..
211* .. Common blocks ..
212 COMMON / infoc / infot, nunit, ok, lerr
213 COMMON / srnamc / srnamt
214* ..
215* .. Intrinsic Functions ..
216 INTRINSIC max, min
217* ..
218* .. Data statements ..
219 DATA iseedy / 1988, 1989, 1990, 1991 /
220 DATA uplos / 'U', 'L' / , facts / 'F', 'N' /
221* ..
222* .. Executable Statements ..
223*
224* Initialize constants and the random number seed.
225*
226* Test path
227*
228 path( 1: 1 ) = 'Double precision'
229 path( 2: 3 ) = 'SK'
230*
231* Path to generate matrices
232*
233 matpath( 1: 1 ) = 'Double precision'
234 matpath( 2: 3 ) = 'SY'
235*
236 nrun = 0
237 nfail = 0
238 nerrs = 0
239 DO 10 i = 1, 4
240 iseed( i ) = iseedy( i )
241 10 CONTINUE
242 lwork = max( 2*nmax, nmax*nrhs )
243*
244* Test the error exits
245*
246 IF( tsterr )
247 \$ CALL derrvx( path, nout )
248 infot = 0
249*
250* Set the block size and minimum block size for which the block
251* routine should be used, which will be later returned by ILAENV.
252*
253 nb = 1
254 nbmin = 2
255 CALL xlaenv( 1, nb )
256 CALL xlaenv( 2, nbmin )
257*
258* Do for each value of N in NVAL
259*
260 DO 180 in = 1, nn
261 n = nval( in )
262 lda = max( n, 1 )
263 xtype = 'N'
264 nimat = ntypes
265 IF( n.LE.0 )
266 \$ nimat = 1
267*
268 DO 170 imat = 1, nimat
269*
270* Do the tests only if DOTYPE( IMAT ) is true.
271*
272 IF( .NOT.dotype( imat ) )
273 \$ GO TO 170
274*
275* Skip types 3, 4, 5, or 6 if the matrix size is too small.
276*
277 zerot = imat.GE.3 .AND. imat.LE.6
278 IF( zerot .AND. n.LT.imat-2 )
279 \$ GO TO 170
280*
281* Do first for UPLO = 'U', then for UPLO = 'L'
282*
283 DO 160 iuplo = 1, 2
284 uplo = uplos( iuplo )
285*
286* Begin generate the test matrix A.
287*
288* Set up parameters with DLATB4 for the matrix generator
289* based on the type of matrix to be generated.
290*
291 CALL dlatb4( matpath, imat, n, n, TYPE, KL, KU, ANORM,
292 \$ MODE, CNDNUM, DIST )
293*
294* Generate a matrix with DLATMS.
295*
296 srnamt = 'DLATMS'
297 CALL dlatms( n, n, dist, iseed, TYPE, RWORK, MODE,
298 \$ CNDNUM, ANORM, KL, KU, UPLO, A, LDA, WORK,
299 \$ INFO )
300*
301* Check error code from DLATMS and handle error.
302*
303 IF( info.NE.0 ) THEN
304 CALL alaerh( path, 'DLATMS', info, 0, uplo, n, n, -1,
305 \$ -1, -1, imat, nfail, nerrs, nout )
306*
307* Skip all tests for this generated matrix
308*
309 GO TO 160
310 END IF
311*
312* For types 3-6, zero one or more rows and columns of the
313* matrix to test that INFO is returned correctly.
314*
315 IF( zerot ) THEN
316 IF( imat.EQ.3 ) THEN
317 izero = 1
318 ELSE IF( imat.EQ.4 ) THEN
319 izero = n
320 ELSE
321 izero = n / 2 + 1
322 END IF
323*
324 IF( imat.LT.6 ) THEN
325*
326* Set row and column IZERO to zero.
327*
328 IF( iuplo.EQ.1 ) THEN
329 ioff = ( izero-1 )*lda
330 DO 20 i = 1, izero - 1
331 a( ioff+i ) = zero
332 20 CONTINUE
333 ioff = ioff + izero
334 DO 30 i = izero, n
335 a( ioff ) = zero
336 ioff = ioff + lda
337 30 CONTINUE
338 ELSE
339 ioff = izero
340 DO 40 i = 1, izero - 1
341 a( ioff ) = zero
342 ioff = ioff + lda
343 40 CONTINUE
344 ioff = ioff - izero
345 DO 50 i = izero, n
346 a( ioff+i ) = zero
347 50 CONTINUE
348 END IF
349 ELSE
350 ioff = 0
351 IF( iuplo.EQ.1 ) THEN
352*
353* Set the first IZERO rows and columns to zero.
354*
355 DO 70 j = 1, n
356 i2 = min( j, izero )
357 DO 60 i = 1, i2
358 a( ioff+i ) = zero
359 60 CONTINUE
360 ioff = ioff + lda
361 70 CONTINUE
362 ELSE
363*
364* Set the last IZERO rows and columns to zero.
365*
366 DO 90 j = 1, n
367 i1 = max( j, izero )
368 DO 80 i = i1, n
369 a( ioff+i ) = zero
370 80 CONTINUE
371 ioff = ioff + lda
372 90 CONTINUE
373 END IF
374 END IF
375 ELSE
376 izero = 0
377 END IF
378*
379* End generate the test matrix A.
380*
381 DO 150 ifact = 1, nfact
382*
383* Do first for FACT = 'F', then for other values.
384*
385 fact = facts( ifact )
386*
387* Compute the condition number
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 = dlansy( '1', uplo, n, a, lda, rwork )
399*
400* Factor the matrix A.
401*
402 CALL dlacpy( uplo, n, n, a, lda, afac, lda )
403 CALL dsytrf_rk( uplo, n, afac, lda, e, iwork, work,
404 \$ lwork, info )
405*
406* Compute inv(A) and take its norm.
407*
408 CALL dlacpy( uplo, n, n, afac, lda, ainv, lda )
409 lwork = (n+nb+1)*(nb+3)
410*
411* We need to compute the inverse to compute
412* RCONDC that is used later in TEST3.
413*
414 CALL dsytri_3( uplo, n, ainv, lda, e, iwork,
415 \$ work, lwork, info )
416 ainvnm = dlansy( '1', uplo, n, ainv, lda, rwork )
417*
418* Compute the 1-norm condition number of A.
419*
420 IF( anorm.LE.zero .OR. ainvnm.LE.zero ) THEN
421 rcondc = one
422 ELSE
423 rcondc = ( one / anorm ) / ainvnm
424 END IF
425 END IF
426*
427* Form an exact solution and set the right hand side.
428*
429 srnamt = 'DLARHS'
430 CALL dlarhs( matpath, xtype, uplo, ' ', n, n, kl, ku,
431 \$ nrhs, a, lda, xact, lda, b, lda, iseed,
432 \$ info )
433 xtype = 'C'
434*
435* --- Test DSYSV_RK ---
436*
437 IF( ifact.EQ.2 ) THEN
438 CALL dlacpy( uplo, n, n, a, lda, afac, lda )
439 CALL dlacpy( 'Full', n, nrhs, b, lda, x, lda )
440*
441* Factor the matrix and solve the system using
442* DSYSV_RK.
443*
444 srnamt = 'DSYSV_RK'
445 CALL dsysv_rk( uplo, n, nrhs, afac, lda, e, iwork,
446 \$ x, lda, work, lwork, info )
447*
448* Adjust the expected value of INFO to account for
449* pivoting.
450*
451 k = izero
452 IF( k.GT.0 ) THEN
453 100 CONTINUE
454 IF( iwork( k ).LT.0 ) THEN
455 IF( iwork( k ).NE.-k ) THEN
456 k = -iwork( k )
457 GO TO 100
458 END IF
459 ELSE IF( iwork( k ).NE.k ) THEN
460 k = iwork( k )
461 GO TO 100
462 END IF
463 END IF
464*
465* Check error code from DSYSV_RK and handle error.
466*
467 IF( info.NE.k ) THEN
468 CALL alaerh( path, 'DSYSV_RK', info, k, uplo,
469 \$ n, n, -1, -1, nrhs, imat, nfail,
470 \$ nerrs, nout )
471 GO TO 120
472 ELSE IF( info.NE.0 ) THEN
473 GO TO 120
474 END IF
475*
476*+ TEST 1 Reconstruct matrix from factors and compute
477* residual.
478*
479 CALL dsyt01_3( uplo, n, a, lda, afac, lda, e,
480 \$ iwork, ainv, lda, rwork,
481 \$ result( 1 ) )
482*
483*+ TEST 2 Compute residual of the computed solution.
484*
485 CALL dlacpy( 'Full', n, nrhs, b, lda, work, lda )
486 CALL dpot02( uplo, n, nrhs, a, lda, x, lda, work,
487 \$ lda, rwork, result( 2 ) )
488*
489*+ TEST 3
490* Check solution from generated exact solution.
491*
492 CALL dget04( n, nrhs, x, lda, xact, lda, rcondc,
493 \$ result( 3 ) )
494 nt = 3
495*
496* Print information about the tests that did not pass
497* the threshold.
498*
499 DO 110 k = 1, nt
500 IF( result( k ).GE.thresh ) THEN
501 IF( nfail.EQ.0 .AND. nerrs.EQ.0 )
502 \$ CALL aladhd( nout, path )
503 WRITE( nout, fmt = 9999 )'DSYSV_RK', uplo,
504 \$ n, imat, k, result( k )
505 nfail = nfail + 1
506 END IF
507 110 CONTINUE
508 nrun = nrun + nt
509 120 CONTINUE
510 END IF
511*
512 150 CONTINUE
513*
514 160 CONTINUE
515 170 CONTINUE
516 180 CONTINUE
517*
518* Print a summary of the results.
519*
520 CALL alasvm( path, nout, nfail, nrun, nerrs )
521*
522 9999 FORMAT( 1x, a, ', UPLO=''', a1, ''', N =', i5, ', type ', i2,
523 \$ ', test ', i2, ', ratio =', g12.5 )
524 RETURN
525*
526* End of DDRVSY_RK
527*
subroutine dlacpy(UPLO, M, N, A, LDA, B, LDB)
DLACPY copies all or part of one two-dimensional array to another.
Definition: dlacpy.f:103
subroutine alasvm(TYPE, NOUT, NFAIL, NRUN, NERRS)
ALASVM
Definition: alasvm.f:73
subroutine xlaenv(ISPEC, NVALUE)
XLAENV
Definition: xlaenv.f:81
subroutine alaerh(PATH, SUBNAM, INFO, INFOE, OPTS, M, N, KL, KU, N5, IMAT, NFAIL, NERRS, NOUT)
ALAERH
Definition: alaerh.f:147
subroutine dlarhs(PATH, XTYPE, UPLO, TRANS, M, N, KL, KU, NRHS, A, LDA, X, LDX, B, LDB, ISEED, INFO)
DLARHS
Definition: dlarhs.f:205
subroutine dsyt01_3(UPLO, N, A, LDA, AFAC, LDAFAC, E, IPIV, C, LDC, RWORK, RESID)
DSYT01_3
Definition: dsyt01_3.f:140
subroutine dget04(N, NRHS, X, LDX, XACT, LDXACT, RCOND, RESID)
DGET04
Definition: dget04.f:102
subroutine dpot02(UPLO, N, NRHS, A, LDA, X, LDX, B, LDB, RWORK, RESID)
DPOT02
Definition: dpot02.f:127
subroutine dlatb4(PATH, IMAT, M, N, TYPE, KL, KU, ANORM, MODE, CNDNUM, DIST)
DLATB4
Definition: dlatb4.f:120
subroutine derrvx(PATH, NUNIT)
DERRVX
Definition: derrvx.f:55
subroutine dlatms(M, N, DIST, ISEED, SYM, D, MODE, COND, DMAX, KL, KU, PACK, A, LDA, WORK, INFO)
DLATMS
Definition: dlatms.f:321
double precision function dlansy(NORM, UPLO, N, A, LDA, WORK)
DLANSY returns the value of the 1-norm, or the Frobenius norm, or the infinity norm,...
Definition: dlansy.f:122
subroutine dsytrf_rk(UPLO, N, A, LDA, E, IPIV, WORK, LWORK, INFO)
DSYTRF_RK computes the factorization of a real symmetric indefinite matrix using the bounded Bunch-Ka...
Definition: dsytrf_rk.f:259
subroutine dsytri_3(UPLO, N, A, LDA, E, IPIV, WORK, LWORK, INFO)
DSYTRI_3
Definition: dsytri_3.f:170
subroutine dsysv_rk(UPLO, N, NRHS, A, LDA, E, IPIV, B, LDB, WORK, LWORK, INFO)
DSYSV_RK computes the solution to system of linear equations A * X = B for SY matrices
Definition: dsysv_rk.f:228
Here is the call graph for this function:
Here is the caller graph for this function: