LAPACK  3.4.1
LAPACK: Linear Algebra PACKage
sdrvvx.f
Go to the documentation of this file.
00001 *> \brief \b SDRVVX
00002 *
00003 *  =========== DOCUMENTATION ===========
00004 *
00005 * Online html documentation available at 
00006 *            http://www.netlib.org/lapack/explore-html/ 
00007 *
00008 *  Definition:
00009 *  ===========
00010 *
00011 *       SUBROUTINE SDRVVX( NSIZES, NN, NTYPES, DOTYPE, ISEED, THRESH,
00012 *                          NIUNIT, NOUNIT, A, LDA, H, WR, WI, WR1, WI1,
00013 *                          VL, LDVL, VR, LDVR, LRE, LDLRE, RCONDV, RCNDV1,
00014 *                          RCDVIN, RCONDE, RCNDE1, RCDEIN, SCALE, SCALE1,
00015 *                          RESULT, WORK, NWORK, IWORK, INFO )
00016 * 
00017 *       .. Scalar Arguments ..
00018 *       INTEGER            INFO, LDA, LDLRE, LDVL, LDVR, NIUNIT, NOUNIT,
00019 *      $                   NSIZES, NTYPES, NWORK
00020 *       REAL               THRESH
00021 *       ..
00022 *       .. Array Arguments ..
00023 *       LOGICAL            DOTYPE( * )
00024 *       INTEGER            ISEED( 4 ), IWORK( * ), NN( * )
00025 *       REAL               A( LDA, * ), H( LDA, * ), LRE( LDLRE, * ),
00026 *      $                   RCDEIN( * ), RCDVIN( * ), RCNDE1( * ),
00027 *      $                   RCNDV1( * ), RCONDE( * ), RCONDV( * ),
00028 *      $                   RESULT( 11 ), SCALE( * ), SCALE1( * ),
00029 *      $                   VL( LDVL, * ), VR( LDVR, * ), WI( * ),
00030 *      $                   WI1( * ), WORK( * ), WR( * ), WR1( * )
00031 *       ..
00032 *  
00033 *
00034 *> \par Purpose:
00035 *  =============
00036 *>
00037 *> \verbatim
00038 *>
00039 *>    SDRVVX  checks the nonsymmetric eigenvalue problem expert driver
00040 *>    SGEEVX.
00041 *>
00042 *>    SDRVVX uses both test matrices generated randomly depending on
00043 *>    data supplied in the calling sequence, as well as on data
00044 *>    read from an input file and including precomputed condition
00045 *>    numbers to which it compares the ones it computes.
00046 *>
00047 *>    When SDRVVX is called, a number of matrix "sizes" ("n's") and a
00048 *>    number of matrix "types" are specified in the calling sequence.
00049 *>    For each size ("n") and each type of matrix, one matrix will be
00050 *>    generated and used to test the nonsymmetric eigenroutines.  For
00051 *>    each matrix, 9 tests will be performed:
00052 *>
00053 *>    (1)     | A * VR - VR * W | / ( n |A| ulp )
00054 *>
00055 *>      Here VR is the matrix of unit right eigenvectors.
00056 *>      W is a block diagonal matrix, with a 1x1 block for each
00057 *>      real eigenvalue and a 2x2 block for each complex conjugate
00058 *>      pair.  If eigenvalues j and j+1 are a complex conjugate pair,
00059 *>      so WR(j) = WR(j+1) = wr and WI(j) = - WI(j+1) = wi, then the
00060 *>      2 x 2 block corresponding to the pair will be:
00061 *>
00062 *>              (  wr  wi  )
00063 *>              ( -wi  wr  )
00064 *>
00065 *>      Such a block multiplying an n x 2 matrix  ( ur ui ) on the
00066 *>      right will be the same as multiplying  ur + i*ui  by  wr + i*wi.
00067 *>
00068 *>    (2)     | A**H * VL - VL * W**H | / ( n |A| ulp )
00069 *>
00070 *>      Here VL is the matrix of unit left eigenvectors, A**H is the
00071 *>      conjugate transpose of A, and W is as above.
00072 *>
00073 *>    (3)     | |VR(i)| - 1 | / ulp and largest component real
00074 *>
00075 *>      VR(i) denotes the i-th column of VR.
00076 *>
00077 *>    (4)     | |VL(i)| - 1 | / ulp and largest component real
00078 *>
00079 *>      VL(i) denotes the i-th column of VL.
00080 *>
00081 *>    (5)     W(full) = W(partial)
00082 *>
00083 *>      W(full) denotes the eigenvalues computed when VR, VL, RCONDV
00084 *>      and RCONDE are also computed, and W(partial) denotes the
00085 *>      eigenvalues computed when only some of VR, VL, RCONDV, and
00086 *>      RCONDE are computed.
00087 *>
00088 *>    (6)     VR(full) = VR(partial)
00089 *>
00090 *>      VR(full) denotes the right eigenvectors computed when VL, RCONDV
00091 *>      and RCONDE are computed, and VR(partial) denotes the result
00092 *>      when only some of VL and RCONDV are computed.
00093 *>
00094 *>    (7)     VL(full) = VL(partial)
00095 *>
00096 *>      VL(full) denotes the left eigenvectors computed when VR, RCONDV
00097 *>      and RCONDE are computed, and VL(partial) denotes the result
00098 *>      when only some of VR and RCONDV are computed.
00099 *>
00100 *>    (8)     0 if SCALE, ILO, IHI, ABNRM (full) =
00101 *>                 SCALE, ILO, IHI, ABNRM (partial)
00102 *>            1/ulp otherwise
00103 *>
00104 *>      SCALE, ILO, IHI and ABNRM describe how the matrix is balanced.
00105 *>      (full) is when VR, VL, RCONDE and RCONDV are also computed, and
00106 *>      (partial) is when some are not computed.
00107 *>
00108 *>    (9)     RCONDV(full) = RCONDV(partial)
00109 *>
00110 *>      RCONDV(full) denotes the reciprocal condition numbers of the
00111 *>      right eigenvectors computed when VR, VL and RCONDE are also
00112 *>      computed. RCONDV(partial) denotes the reciprocal condition
00113 *>      numbers when only some of VR, VL and RCONDE are computed.
00114 *>
00115 *>    The "sizes" are specified by an array NN(1:NSIZES); the value of
00116 *>    each element NN(j) specifies one size.
00117 *>    The "types" are specified by a logical array DOTYPE( 1:NTYPES );
00118 *>    if DOTYPE(j) is .TRUE., then matrix type "j" will be generated.
00119 *>    Currently, the list of possible types is:
00120 *>
00121 *>    (1)  The zero matrix.
00122 *>    (2)  The identity matrix.
00123 *>    (3)  A (transposed) Jordan block, with 1's on the diagonal.
00124 *>
00125 *>    (4)  A diagonal matrix with evenly spaced entries
00126 *>         1, ..., ULP  and random signs.
00127 *>         (ULP = (first number larger than 1) - 1 )
00128 *>    (5)  A diagonal matrix with geometrically spaced entries
00129 *>         1, ..., ULP  and random signs.
00130 *>    (6)  A diagonal matrix with "clustered" entries 1, ULP, ..., ULP
00131 *>         and random signs.
00132 *>
00133 *>    (7)  Same as (4), but multiplied by a constant near
00134 *>         the overflow threshold
00135 *>    (8)  Same as (4), but multiplied by a constant near
00136 *>         the underflow threshold
00137 *>
00138 *>    (9)  A matrix of the form  U' T U, where U is orthogonal and
00139 *>         T has evenly spaced entries 1, ..., ULP with random signs
00140 *>         on the diagonal and random O(1) entries in the upper
00141 *>         triangle.
00142 *>
00143 *>    (10) A matrix of the form  U' T U, where U is orthogonal and
00144 *>         T has geometrically spaced entries 1, ..., ULP with random
00145 *>         signs on the diagonal and random O(1) entries in the upper
00146 *>         triangle.
00147 *>
00148 *>    (11) A matrix of the form  U' T U, where U is orthogonal and
00149 *>         T has "clustered" entries 1, ULP,..., ULP with random
00150 *>         signs on the diagonal and random O(1) entries in the upper
00151 *>         triangle.
00152 *>
00153 *>    (12) A matrix of the form  U' T U, where U is orthogonal and
00154 *>         T has real or complex conjugate paired eigenvalues randomly
00155 *>         chosen from ( ULP, 1 ) and random O(1) entries in the upper
00156 *>         triangle.
00157 *>
00158 *>    (13) A matrix of the form  X' T X, where X has condition
00159 *>         SQRT( ULP ) and T has evenly spaced entries 1, ..., ULP
00160 *>         with random signs on the diagonal and random O(1) entries
00161 *>         in the upper triangle.
00162 *>
00163 *>    (14) A matrix of the form  X' T X, where X has condition
00164 *>         SQRT( ULP ) and T has geometrically spaced entries
00165 *>         1, ..., ULP with random signs on the diagonal and random
00166 *>         O(1) entries in the upper triangle.
00167 *>
00168 *>    (15) A matrix of the form  X' T X, where X has condition
00169 *>         SQRT( ULP ) and T has "clustered" entries 1, ULP,..., ULP
00170 *>         with random signs on the diagonal and random O(1) entries
00171 *>         in the upper triangle.
00172 *>
00173 *>    (16) A matrix of the form  X' T X, where X has condition
00174 *>         SQRT( ULP ) and T has real or complex conjugate paired
00175 *>         eigenvalues randomly chosen from ( ULP, 1 ) and random
00176 *>         O(1) entries in the upper triangle.
00177 *>
00178 *>    (17) Same as (16), but multiplied by a constant
00179 *>         near the overflow threshold
00180 *>    (18) Same as (16), but multiplied by a constant
00181 *>         near the underflow threshold
00182 *>
00183 *>    (19) Nonsymmetric matrix with random entries chosen from (-1,1).
00184 *>         If N is at least 4, all entries in first two rows and last
00185 *>         row, and first column and last two columns are zero.
00186 *>    (20) Same as (19), but multiplied by a constant
00187 *>         near the overflow threshold
00188 *>    (21) Same as (19), but multiplied by a constant
00189 *>         near the underflow threshold
00190 *>
00191 *>    In addition, an input file will be read from logical unit number
00192 *>    NIUNIT. The file contains matrices along with precomputed
00193 *>    eigenvalues and reciprocal condition numbers for the eigenvalues
00194 *>    and right eigenvectors. For these matrices, in addition to tests
00195 *>    (1) to (9) we will compute the following two tests:
00196 *>
00197 *>   (10)  |RCONDV - RCDVIN| / cond(RCONDV)
00198 *>
00199 *>      RCONDV is the reciprocal right eigenvector condition number
00200 *>      computed by SGEEVX and RCDVIN (the precomputed true value)
00201 *>      is supplied as input. cond(RCONDV) is the condition number of
00202 *>      RCONDV, and takes errors in computing RCONDV into account, so
00203 *>      that the resulting quantity should be O(ULP). cond(RCONDV) is
00204 *>      essentially given by norm(A)/RCONDE.
00205 *>
00206 *>   (11)  |RCONDE - RCDEIN| / cond(RCONDE)
00207 *>
00208 *>      RCONDE is the reciprocal eigenvalue condition number
00209 *>      computed by SGEEVX and RCDEIN (the precomputed true value)
00210 *>      is supplied as input.  cond(RCONDE) is the condition number
00211 *>      of RCONDE, and takes errors in computing RCONDE into account,
00212 *>      so that the resulting quantity should be O(ULP). cond(RCONDE)
00213 *>      is essentially given by norm(A)/RCONDV.
00214 *> \endverbatim
00215 *
00216 *  Arguments:
00217 *  ==========
00218 *
00219 *> \param[in] NSIZES
00220 *> \verbatim
00221 *>          NSIZES is INTEGER
00222 *>          The number of sizes of matrices to use.  NSIZES must be at
00223 *>          least zero. If it is zero, no randomly generated matrices
00224 *>          are tested, but any test matrices read from NIUNIT will be
00225 *>          tested.
00226 *> \endverbatim
00227 *>
00228 *> \param[in] NN
00229 *> \verbatim
00230 *>          NN is INTEGER array, dimension (NSIZES)
00231 *>          An array containing the sizes to be used for the matrices.
00232 *>          Zero values will be skipped.  The values must be at least
00233 *>          zero.
00234 *> \endverbatim
00235 *>
00236 *> \param[in] NTYPES
00237 *> \verbatim
00238 *>          NTYPES is INTEGER
00239 *>          The number of elements in DOTYPE. NTYPES must be at least
00240 *>          zero. If it is zero, no randomly generated test matrices
00241 *>          are tested, but and test matrices read from NIUNIT will be
00242 *>          tested. If it is MAXTYP+1 and NSIZES is 1, then an
00243 *>          additional type, MAXTYP+1 is defined, which is to use
00244 *>          whatever matrix is in A.  This is only useful if
00245 *>          DOTYPE(1:MAXTYP) is .FALSE. and DOTYPE(MAXTYP+1) is .TRUE. .
00246 *> \endverbatim
00247 *>
00248 *> \param[in] DOTYPE
00249 *> \verbatim
00250 *>          DOTYPE is LOGICAL array, dimension (NTYPES)
00251 *>          If DOTYPE(j) is .TRUE., then for each size in NN a
00252 *>          matrix of that size and of type j will be generated.
00253 *>          If NTYPES is smaller than the maximum number of types
00254 *>          defined (PARAMETER MAXTYP), then types NTYPES+1 through
00255 *>          MAXTYP will not be generated.  If NTYPES is larger
00256 *>          than MAXTYP, DOTYPE(MAXTYP+1) through DOTYPE(NTYPES)
00257 *>          will be ignored.
00258 *> \endverbatim
00259 *>
00260 *> \param[in,out] ISEED
00261 *> \verbatim
00262 *>          ISEED is INTEGER array, dimension (4)
00263 *>          On entry ISEED specifies the seed of the random number
00264 *>          generator. The array elements should be between 0 and 4095;
00265 *>          if not they will be reduced mod 4096.  Also, ISEED(4) must
00266 *>          be odd.  The random number generator uses a linear
00267 *>          congruential sequence limited to small integers, and so
00268 *>          should produce machine independent random numbers. The
00269 *>          values of ISEED are changed on exit, and can be used in the
00270 *>          next call to SDRVVX to continue the same random number
00271 *>          sequence.
00272 *> \endverbatim
00273 *>
00274 *> \param[in] THRESH
00275 *> \verbatim
00276 *>          THRESH is REAL
00277 *>          A test will count as "failed" if the "error", computed as
00278 *>          described above, exceeds THRESH.  Note that the error
00279 *>          is scaled to be O(1), so THRESH should be a reasonably
00280 *>          small multiple of 1, e.g., 10 or 100.  In particular,
00281 *>          it should not depend on the precision (single vs. double)
00282 *>          or the size of the matrix.  It must be at least zero.
00283 *> \endverbatim
00284 *>
00285 *> \param[in] NIUNIT
00286 *> \verbatim
00287 *>          NIUNIT is INTEGER
00288 *>          The FORTRAN unit number for reading in the data file of
00289 *>          problems to solve.
00290 *> \endverbatim
00291 *>
00292 *> \param[in] NOUNIT
00293 *> \verbatim
00294 *>          NOUNIT is INTEGER
00295 *>          The FORTRAN unit number for printing out error messages
00296 *>          (e.g., if a routine returns INFO not equal to 0.)
00297 *> \endverbatim
00298 *>
00299 *> \param[out] A
00300 *> \verbatim
00301 *>          A is REAL array, dimension
00302 *>                      (LDA, max(NN,12))
00303 *>          Used to hold the matrix whose eigenvalues are to be
00304 *>          computed.  On exit, A contains the last matrix actually used.
00305 *> \endverbatim
00306 *>
00307 *> \param[in] LDA
00308 *> \verbatim
00309 *>          LDA is INTEGER
00310 *>          The leading dimension of the arrays A and H.
00311 *>          LDA >= max(NN,12), since 12 is the dimension of the largest
00312 *>          matrix in the precomputed input file.
00313 *> \endverbatim
00314 *>
00315 *> \param[out] H
00316 *> \verbatim
00317 *>          H is REAL array, dimension
00318 *>                      (LDA, max(NN,12))
00319 *>          Another copy of the test matrix A, modified by SGEEVX.
00320 *> \endverbatim
00321 *>
00322 *> \param[out] WR
00323 *> \verbatim
00324 *>          WR is REAL array, dimension (max(NN))
00325 *> \endverbatim
00326 *>
00327 *> \param[out] WI
00328 *> \verbatim
00329 *>          WI is REAL array, dimension (max(NN))
00330 *>          The real and imaginary parts of the eigenvalues of A.
00331 *>          On exit, WR + WI*i are the eigenvalues of the matrix in A.
00332 *> \endverbatim
00333 *>
00334 *> \param[out] WR1
00335 *> \verbatim
00336 *>          WR1 is REAL array, dimension (max(NN,12))
00337 *> \endverbatim
00338 *>
00339 *> \param[out] WI1
00340 *> \verbatim
00341 *>          WI1 is REAL array, dimension (max(NN,12))
00342 *>
00343 *>          Like WR, WI, these arrays contain the eigenvalues of A,
00344 *>          but those computed when SGEEVX only computes a partial
00345 *>          eigendecomposition, i.e. not the eigenvalues and left
00346 *>          and right eigenvectors.
00347 *> \endverbatim
00348 *>
00349 *> \param[out] VL
00350 *> \verbatim
00351 *>          VL is REAL array, dimension
00352 *>                      (LDVL, max(NN,12))
00353 *>          VL holds the computed left eigenvectors.
00354 *> \endverbatim
00355 *>
00356 *> \param[in] LDVL
00357 *> \verbatim
00358 *>          LDVL is INTEGER
00359 *>          Leading dimension of VL. Must be at least max(1,max(NN,12)).
00360 *> \endverbatim
00361 *>
00362 *> \param[out] VR
00363 *> \verbatim
00364 *>          VR is REAL array, dimension
00365 *>                      (LDVR, max(NN,12))
00366 *>          VR holds the computed right eigenvectors.
00367 *> \endverbatim
00368 *>
00369 *> \param[in] LDVR
00370 *> \verbatim
00371 *>          LDVR is INTEGER
00372 *>          Leading dimension of VR. Must be at least max(1,max(NN,12)).
00373 *> \endverbatim
00374 *>
00375 *> \param[out] LRE
00376 *> \verbatim
00377 *>          LRE is REAL array, dimension
00378 *>                      (LDLRE, max(NN,12))
00379 *>          LRE holds the computed right or left eigenvectors.
00380 *> \endverbatim
00381 *>
00382 *> \param[in] LDLRE
00383 *> \verbatim
00384 *>          LDLRE is INTEGER
00385 *>          Leading dimension of LRE. Must be at least max(1,max(NN,12))
00386 *> \endverbatim
00387 *>
00388 *> \param[out] RCONDV
00389 *> \verbatim
00390 *>          RCONDV is REAL array, dimension (N)
00391 *>          RCONDV holds the computed reciprocal condition numbers
00392 *>          for eigenvectors.
00393 *> \endverbatim
00394 *>
00395 *> \param[out] RCNDV1
00396 *> \verbatim
00397 *>          RCNDV1 is REAL array, dimension (N)
00398 *>          RCNDV1 holds more computed reciprocal condition numbers
00399 *>          for eigenvectors.
00400 *> \endverbatim
00401 *>
00402 *> \param[out] RCDVIN
00403 *> \verbatim
00404 *>          RCDVIN is REAL array, dimension (N)
00405 *>          When COMP = .TRUE. RCDVIN holds the precomputed reciprocal
00406 *>          condition numbers for eigenvectors to be compared with
00407 *>          RCONDV.
00408 *> \endverbatim
00409 *>
00410 *> \param[out] RCONDE
00411 *> \verbatim
00412 *>          RCONDE is REAL array, dimension (N)
00413 *>          RCONDE holds the computed reciprocal condition numbers
00414 *>          for eigenvalues.
00415 *> \endverbatim
00416 *>
00417 *> \param[out] RCNDE1
00418 *> \verbatim
00419 *>          RCNDE1 is REAL array, dimension (N)
00420 *>          RCNDE1 holds more computed reciprocal condition numbers
00421 *>          for eigenvalues.
00422 *> \endverbatim
00423 *>
00424 *> \param[out] RCDEIN
00425 *> \verbatim
00426 *>          RCDEIN is REAL array, dimension (N)
00427 *>          When COMP = .TRUE. RCDEIN holds the precomputed reciprocal
00428 *>          condition numbers for eigenvalues to be compared with
00429 *>          RCONDE.
00430 *> \endverbatim
00431 *>
00432 *> \param[out] SCALE
00433 *> \verbatim
00434 *>          SCALE is REAL array, dimension (N)
00435 *>          Holds information describing balancing of matrix.
00436 *> \endverbatim
00437 *>
00438 *> \param[out] SCALE1
00439 *> \verbatim
00440 *>          SCALE1 is REAL array, dimension (N)
00441 *>          Holds information describing balancing of matrix.
00442 *> \endverbatim
00443 *>
00444 *> \param[out] RESULT
00445 *> \verbatim
00446 *>          RESULT is REAL array, dimension (11)
00447 *>          The values computed by the seven tests described above.
00448 *>          The values are currently limited to 1/ulp, to avoid overflow.
00449 *> \endverbatim
00450 *>
00451 *> \param[out] WORK
00452 *> \verbatim
00453 *>          WORK is REAL array, dimension (NWORK)
00454 *> \endverbatim
00455 *>
00456 *> \param[in] NWORK
00457 *> \verbatim
00458 *>          NWORK is INTEGER
00459 *>          The number of entries in WORK.  This must be at least
00460 *>          max(6*12+2*12**2,6*NN(j)+2*NN(j)**2) =
00461 *>          max(    360     ,6*NN(j)+2*NN(j)**2)    for all j.
00462 *> \endverbatim
00463 *>
00464 *> \param[out] IWORK
00465 *> \verbatim
00466 *>          IWORK is INTEGER array, dimension (2*max(NN,12))
00467 *> \endverbatim
00468 *>
00469 *> \param[out] INFO
00470 *> \verbatim
00471 *>          INFO is INTEGER
00472 *>          If 0,  then successful exit.
00473 *>          If <0, then input paramter -INFO is incorrect.
00474 *>          If >0, SLATMR, SLATMS, SLATME or SGET23 returned an error
00475 *>                 code, and INFO is its absolute value.
00476 *>
00477 *>-----------------------------------------------------------------------
00478 *>
00479 *>     Some Local Variables and Parameters:
00480 *>     ---- ----- --------- --- ----------
00481 *>
00482 *>     ZERO, ONE       Real 0 and 1.
00483 *>     MAXTYP          The number of types defined.
00484 *>     NMAX            Largest value in NN or 12.
00485 *>     NERRS           The number of tests which have exceeded THRESH
00486 *>     COND, CONDS,
00487 *>     IMODE           Values to be passed to the matrix generators.
00488 *>     ANORM           Norm of A; passed to matrix generators.
00489 *>
00490 *>     OVFL, UNFL      Overflow and underflow thresholds.
00491 *>     ULP, ULPINV     Finest relative precision and its inverse.
00492 *>     RTULP, RTULPI   Square roots of the previous 4 values.
00493 *>
00494 *>             The following four arrays decode JTYPE:
00495 *>     KTYPE(j)        The general type (1-10) for type "j".
00496 *>     KMODE(j)        The MODE value to be passed to the matrix
00497 *>                     generator for type "j".
00498 *>     KMAGN(j)        The order of magnitude ( O(1),
00499 *>                     O(overflow^(1/2) ), O(underflow^(1/2) )
00500 *>     KCONDS(j)       Selectw whether CONDS is to be 1 or
00501 *>                     1/sqrt(ulp).  (0 means irrelevant.)
00502 *> \endverbatim
00503 *
00504 *  Authors:
00505 *  ========
00506 *
00507 *> \author Univ. of Tennessee 
00508 *> \author Univ. of California Berkeley 
00509 *> \author Univ. of Colorado Denver 
00510 *> \author NAG Ltd. 
00511 *
00512 *> \date November 2011
00513 *
00514 *> \ingroup single_eig
00515 *
00516 *  =====================================================================
00517       SUBROUTINE SDRVVX( NSIZES, NN, NTYPES, DOTYPE, ISEED, THRESH,
00518      $                   NIUNIT, NOUNIT, A, LDA, H, WR, WI, WR1, WI1,
00519      $                   VL, LDVL, VR, LDVR, LRE, LDLRE, RCONDV, RCNDV1,
00520      $                   RCDVIN, RCONDE, RCNDE1, RCDEIN, SCALE, SCALE1,
00521      $                   RESULT, WORK, NWORK, IWORK, INFO )
00522 *
00523 *  -- LAPACK test routine (version 3.4.0) --
00524 *  -- LAPACK is a software package provided by Univ. of Tennessee,    --
00525 *  -- Univ. of California Berkeley, Univ. of Colorado Denver and NAG Ltd..--
00526 *     November 2011
00527 *
00528 *     .. Scalar Arguments ..
00529       INTEGER            INFO, LDA, LDLRE, LDVL, LDVR, NIUNIT, NOUNIT,
00530      $                   NSIZES, NTYPES, NWORK
00531       REAL               THRESH
00532 *     ..
00533 *     .. Array Arguments ..
00534       LOGICAL            DOTYPE( * )
00535       INTEGER            ISEED( 4 ), IWORK( * ), NN( * )
00536       REAL               A( LDA, * ), H( LDA, * ), LRE( LDLRE, * ),
00537      $                   RCDEIN( * ), RCDVIN( * ), RCNDE1( * ),
00538      $                   RCNDV1( * ), RCONDE( * ), RCONDV( * ),
00539      $                   RESULT( 11 ), SCALE( * ), SCALE1( * ),
00540      $                   VL( LDVL, * ), VR( LDVR, * ), WI( * ),
00541      $                   WI1( * ), WORK( * ), WR( * ), WR1( * )
00542 *     ..
00543 *
00544 *  =====================================================================
00545 *
00546 *     .. Parameters ..
00547       REAL               ZERO, ONE
00548       PARAMETER          ( ZERO = 0.0E0, ONE = 1.0E0 )
00549       INTEGER            MAXTYP
00550       PARAMETER          ( MAXTYP = 21 )
00551 *     ..
00552 *     .. Local Scalars ..
00553       LOGICAL            BADNN
00554       CHARACTER          BALANC
00555       CHARACTER*3        PATH
00556       INTEGER            I, IBAL, IINFO, IMODE, ITYPE, IWK, J, JCOL,
00557      $                   JSIZE, JTYPE, MTYPES, N, NERRS, NFAIL,
00558      $                   NMAX, NNWORK, NTEST, NTESTF, NTESTT
00559       REAL               ANORM, COND, CONDS, OVFL, RTULP, RTULPI, ULP,
00560      $                   ULPINV, UNFL
00561 *     ..
00562 *     .. Local Arrays ..
00563       CHARACTER          ADUMMA( 1 ), BAL( 4 )
00564       INTEGER            IDUMMA( 1 ), IOLDSD( 4 ), KCONDS( MAXTYP ),
00565      $                   KMAGN( MAXTYP ), KMODE( MAXTYP ),
00566      $                   KTYPE( MAXTYP )
00567 *     ..
00568 *     .. External Functions ..
00569       REAL               SLAMCH
00570       EXTERNAL           SLAMCH
00571 *     ..
00572 *     .. External Subroutines ..
00573       EXTERNAL           SGET23, SLABAD, SLASUM, SLATME, SLATMR, SLATMS,
00574      $                   SLASET, XERBLA
00575 *     ..
00576 *     .. Intrinsic Functions ..
00577       INTRINSIC          ABS, MAX, MIN, SQRT
00578 *     ..
00579 *     .. Data statements ..
00580       DATA               KTYPE / 1, 2, 3, 5*4, 4*6, 6*6, 3*9 /
00581       DATA               KMAGN / 3*1, 1, 1, 1, 2, 3, 4*1, 1, 1, 1, 1, 2,
00582      $                   3, 1, 2, 3 /
00583       DATA               KMODE / 3*0, 4, 3, 1, 4, 4, 4, 3, 1, 5, 4, 3,
00584      $                   1, 5, 5, 5, 4, 3, 1 /
00585       DATA               KCONDS / 3*0, 5*0, 4*1, 6*2, 3*0 /
00586       DATA               BAL / 'N', 'P', 'S', 'B' /
00587 *     ..
00588 *     .. Executable Statements ..
00589 *
00590       PATH( 1: 1 ) = 'Single precision'
00591       PATH( 2: 3 ) = 'VX'
00592 *
00593 *     Check for errors
00594 *
00595       NTESTT = 0
00596       NTESTF = 0
00597       INFO = 0
00598 *
00599 *     Important constants
00600 *
00601       BADNN = .FALSE.
00602 *
00603 *     12 is the largest dimension in the input file of precomputed
00604 *     problems
00605 *
00606       NMAX = 12
00607       DO 10 J = 1, NSIZES
00608          NMAX = MAX( NMAX, NN( J ) )
00609          IF( NN( J ).LT.0 )
00610      $      BADNN = .TRUE.
00611    10 CONTINUE
00612 *
00613 *     Check for errors
00614 *
00615       IF( NSIZES.LT.0 ) THEN
00616          INFO = -1
00617       ELSE IF( BADNN ) THEN
00618          INFO = -2
00619       ELSE IF( NTYPES.LT.0 ) THEN
00620          INFO = -3
00621       ELSE IF( THRESH.LT.ZERO ) THEN
00622          INFO = -6
00623       ELSE IF( LDA.LT.1 .OR. LDA.LT.NMAX ) THEN
00624          INFO = -10
00625       ELSE IF( LDVL.LT.1 .OR. LDVL.LT.NMAX ) THEN
00626          INFO = -17
00627       ELSE IF( LDVR.LT.1 .OR. LDVR.LT.NMAX ) THEN
00628          INFO = -19
00629       ELSE IF( LDLRE.LT.1 .OR. LDLRE.LT.NMAX ) THEN
00630          INFO = -21
00631       ELSE IF( 6*NMAX+2*NMAX**2.GT.NWORK ) THEN
00632          INFO = -32
00633       END IF
00634 *
00635       IF( INFO.NE.0 ) THEN
00636          CALL XERBLA( 'SDRVVX', -INFO )
00637          RETURN
00638       END IF
00639 *
00640 *     If nothing to do check on NIUNIT
00641 *
00642       IF( NSIZES.EQ.0 .OR. NTYPES.EQ.0 )
00643      $   GO TO 160
00644 *
00645 *     More Important constants
00646 *
00647       UNFL = SLAMCH( 'Safe minimum' )
00648       OVFL = ONE / UNFL
00649       CALL SLABAD( UNFL, OVFL )
00650       ULP = SLAMCH( 'Precision' )
00651       ULPINV = ONE / ULP
00652       RTULP = SQRT( ULP )
00653       RTULPI = ONE / RTULP
00654 *
00655 *     Loop over sizes, types
00656 *
00657       NERRS = 0
00658 *
00659       DO 150 JSIZE = 1, NSIZES
00660          N = NN( JSIZE )
00661          IF( NSIZES.NE.1 ) THEN
00662             MTYPES = MIN( MAXTYP, NTYPES )
00663          ELSE
00664             MTYPES = MIN( MAXTYP+1, NTYPES )
00665          END IF
00666 *
00667          DO 140 JTYPE = 1, MTYPES
00668             IF( .NOT.DOTYPE( JTYPE ) )
00669      $         GO TO 140
00670 *
00671 *           Save ISEED in case of an error.
00672 *
00673             DO 20 J = 1, 4
00674                IOLDSD( J ) = ISEED( J )
00675    20       CONTINUE
00676 *
00677 *           Compute "A"
00678 *
00679 *           Control parameters:
00680 *
00681 *           KMAGN  KCONDS  KMODE        KTYPE
00682 *       =1  O(1)   1       clustered 1  zero
00683 *       =2  large  large   clustered 2  identity
00684 *       =3  small          exponential  Jordan
00685 *       =4                 arithmetic   diagonal, (w/ eigenvalues)
00686 *       =5                 random log   symmetric, w/ eigenvalues
00687 *       =6                 random       general, w/ eigenvalues
00688 *       =7                              random diagonal
00689 *       =8                              random symmetric
00690 *       =9                              random general
00691 *       =10                             random triangular
00692 *
00693             IF( MTYPES.GT.MAXTYP )
00694      $         GO TO 90
00695 *
00696             ITYPE = KTYPE( JTYPE )
00697             IMODE = KMODE( JTYPE )
00698 *
00699 *           Compute norm
00700 *
00701             GO TO ( 30, 40, 50 )KMAGN( JTYPE )
00702 *
00703    30       CONTINUE
00704             ANORM = ONE
00705             GO TO 60
00706 *
00707    40       CONTINUE
00708             ANORM = OVFL*ULP
00709             GO TO 60
00710 *
00711    50       CONTINUE
00712             ANORM = UNFL*ULPINV
00713             GO TO 60
00714 *
00715    60       CONTINUE
00716 *
00717             CALL SLASET( 'Full', LDA, N, ZERO, ZERO, A, LDA )
00718             IINFO = 0
00719             COND = ULPINV
00720 *
00721 *           Special Matrices -- Identity & Jordan block
00722 *
00723 *              Zero
00724 *
00725             IF( ITYPE.EQ.1 ) THEN
00726                IINFO = 0
00727 *
00728             ELSE IF( ITYPE.EQ.2 ) THEN
00729 *
00730 *              Identity
00731 *
00732                DO 70 JCOL = 1, N
00733                   A( JCOL, JCOL ) = ANORM
00734    70          CONTINUE
00735 *
00736             ELSE IF( ITYPE.EQ.3 ) THEN
00737 *
00738 *              Jordan Block
00739 *
00740                DO 80 JCOL = 1, N
00741                   A( JCOL, JCOL ) = ANORM
00742                   IF( JCOL.GT.1 )
00743      $               A( JCOL, JCOL-1 ) = ONE
00744    80          CONTINUE
00745 *
00746             ELSE IF( ITYPE.EQ.4 ) THEN
00747 *
00748 *              Diagonal Matrix, [Eigen]values Specified
00749 *
00750                CALL SLATMS( N, N, 'S', ISEED, 'S', WORK, IMODE, COND,
00751      $                      ANORM, 0, 0, 'N', A, LDA, WORK( N+1 ),
00752      $                      IINFO )
00753 *
00754             ELSE IF( ITYPE.EQ.5 ) THEN
00755 *
00756 *              Symmetric, eigenvalues specified
00757 *
00758                CALL SLATMS( N, N, 'S', ISEED, 'S', WORK, IMODE, COND,
00759      $                      ANORM, N, N, 'N', A, LDA, WORK( N+1 ),
00760      $                      IINFO )
00761 *
00762             ELSE IF( ITYPE.EQ.6 ) THEN
00763 *
00764 *              General, eigenvalues specified
00765 *
00766                IF( KCONDS( JTYPE ).EQ.1 ) THEN
00767                   CONDS = ONE
00768                ELSE IF( KCONDS( JTYPE ).EQ.2 ) THEN
00769                   CONDS = RTULPI
00770                ELSE
00771                   CONDS = ZERO
00772                END IF
00773 *
00774                ADUMMA( 1 ) = ' '
00775                CALL SLATME( N, 'S', ISEED, WORK, IMODE, COND, ONE,
00776      $                      ADUMMA, 'T', 'T', 'T', WORK( N+1 ), 4,
00777      $                      CONDS, N, N, ANORM, A, LDA, WORK( 2*N+1 ),
00778      $                      IINFO )
00779 *
00780             ELSE IF( ITYPE.EQ.7 ) THEN
00781 *
00782 *              Diagonal, random eigenvalues
00783 *
00784                CALL SLATMR( N, N, 'S', ISEED, 'S', WORK, 6, ONE, ONE,
00785      $                      'T', 'N', WORK( N+1 ), 1, ONE,
00786      $                      WORK( 2*N+1 ), 1, ONE, 'N', IDUMMA, 0, 0,
00787      $                      ZERO, ANORM, 'NO', A, LDA, IWORK, IINFO )
00788 *
00789             ELSE IF( ITYPE.EQ.8 ) THEN
00790 *
00791 *              Symmetric, random eigenvalues
00792 *
00793                CALL SLATMR( N, N, 'S', ISEED, 'S', WORK, 6, ONE, ONE,
00794      $                      'T', 'N', WORK( N+1 ), 1, ONE,
00795      $                      WORK( 2*N+1 ), 1, ONE, 'N', IDUMMA, N, N,
00796      $                      ZERO, ANORM, 'NO', A, LDA, IWORK, IINFO )
00797 *
00798             ELSE IF( ITYPE.EQ.9 ) THEN
00799 *
00800 *              General, random eigenvalues
00801 *
00802                CALL SLATMR( N, N, 'S', ISEED, 'N', WORK, 6, ONE, ONE,
00803      $                      'T', 'N', WORK( N+1 ), 1, ONE,
00804      $                      WORK( 2*N+1 ), 1, ONE, 'N', IDUMMA, N, N,
00805      $                      ZERO, ANORM, 'NO', A, LDA, IWORK, IINFO )
00806                IF( N.GE.4 ) THEN
00807                   CALL SLASET( 'Full', 2, N, ZERO, ZERO, A, LDA )
00808                   CALL SLASET( 'Full', N-3, 1, ZERO, ZERO, A( 3, 1 ),
00809      $                         LDA )
00810                   CALL SLASET( 'Full', N-3, 2, ZERO, ZERO, A( 3, N-1 ),
00811      $                         LDA )
00812                   CALL SLASET( 'Full', 1, N, ZERO, ZERO, A( N, 1 ),
00813      $                         LDA )
00814                END IF
00815 *
00816             ELSE IF( ITYPE.EQ.10 ) THEN
00817 *
00818 *              Triangular, random eigenvalues
00819 *
00820                CALL SLATMR( N, N, 'S', ISEED, 'N', WORK, 6, ONE, ONE,
00821      $                      'T', 'N', WORK( N+1 ), 1, ONE,
00822      $                      WORK( 2*N+1 ), 1, ONE, 'N', IDUMMA, N, 0,
00823      $                      ZERO, ANORM, 'NO', A, LDA, IWORK, IINFO )
00824 *
00825             ELSE
00826 *
00827                IINFO = 1
00828             END IF
00829 *
00830             IF( IINFO.NE.0 ) THEN
00831                WRITE( NOUNIT, FMT = 9992 )'Generator', IINFO, N, JTYPE,
00832      $            IOLDSD
00833                INFO = ABS( IINFO )
00834                RETURN
00835             END IF
00836 *
00837    90       CONTINUE
00838 *
00839 *           Test for minimal and generous workspace
00840 *
00841             DO 130 IWK = 1, 3
00842                IF( IWK.EQ.1 ) THEN
00843                   NNWORK = 3*N
00844                ELSE IF( IWK.EQ.2 ) THEN
00845                   NNWORK = 6*N + N**2
00846                ELSE
00847                   NNWORK = 6*N + 2*N**2
00848                END IF
00849                NNWORK = MAX( NNWORK, 1 )
00850 *
00851 *              Test for all balancing options
00852 *
00853                DO 120 IBAL = 1, 4
00854                   BALANC = BAL( IBAL )
00855 *
00856 *                 Perform tests
00857 *
00858                   CALL SGET23( .FALSE., BALANC, JTYPE, THRESH, IOLDSD,
00859      $                         NOUNIT, N, A, LDA, H, WR, WI, WR1, WI1,
00860      $                         VL, LDVL, VR, LDVR, LRE, LDLRE, RCONDV,
00861      $                         RCNDV1, RCDVIN, RCONDE, RCNDE1, RCDEIN,
00862      $                         SCALE, SCALE1, RESULT, WORK, NNWORK,
00863      $                         IWORK, INFO )
00864 *
00865 *                 Check for RESULT(j) > THRESH
00866 *
00867                   NTEST = 0
00868                   NFAIL = 0
00869                   DO 100 J = 1, 9
00870                      IF( RESULT( J ).GE.ZERO )
00871      $                  NTEST = NTEST + 1
00872                      IF( RESULT( J ).GE.THRESH )
00873      $                  NFAIL = NFAIL + 1
00874   100             CONTINUE
00875 *
00876                   IF( NFAIL.GT.0 )
00877      $               NTESTF = NTESTF + 1
00878                   IF( NTESTF.EQ.1 ) THEN
00879                      WRITE( NOUNIT, FMT = 9999 )PATH
00880                      WRITE( NOUNIT, FMT = 9998 )
00881                      WRITE( NOUNIT, FMT = 9997 )
00882                      WRITE( NOUNIT, FMT = 9996 )
00883                      WRITE( NOUNIT, FMT = 9995 )THRESH
00884                      NTESTF = 2
00885                   END IF
00886 *
00887                   DO 110 J = 1, 9
00888                      IF( RESULT( J ).GE.THRESH ) THEN
00889                         WRITE( NOUNIT, FMT = 9994 )BALANC, N, IWK,
00890      $                     IOLDSD, JTYPE, J, RESULT( J )
00891                      END IF
00892   110             CONTINUE
00893 *
00894                   NERRS = NERRS + NFAIL
00895                   NTESTT = NTESTT + NTEST
00896 *
00897   120          CONTINUE
00898   130       CONTINUE
00899   140    CONTINUE
00900   150 CONTINUE
00901 *
00902   160 CONTINUE
00903 *
00904 *     Read in data from file to check accuracy of condition estimation.
00905 *     Assume input eigenvalues are sorted lexicographically (increasing
00906 *     by real part, then decreasing by imaginary part)
00907 *
00908       JTYPE = 0
00909   170 CONTINUE
00910       READ( NIUNIT, FMT = *, END = 220 )N
00911 *
00912 *     Read input data until N=0
00913 *
00914       IF( N.EQ.0 )
00915      $   GO TO 220
00916       JTYPE = JTYPE + 1
00917       ISEED( 1 ) = JTYPE
00918       DO 180 I = 1, N
00919          READ( NIUNIT, FMT = * )( A( I, J ), J = 1, N )
00920   180 CONTINUE
00921       DO 190 I = 1, N
00922          READ( NIUNIT, FMT = * )WR1( I ), WI1( I ), RCDEIN( I ),
00923      $      RCDVIN( I )
00924   190 CONTINUE
00925       CALL SGET23( .TRUE., 'N', 22, THRESH, ISEED, NOUNIT, N, A, LDA, H,
00926      $             WR, WI, WR1, WI1, VL, LDVL, VR, LDVR, LRE, LDLRE,
00927      $             RCONDV, RCNDV1, RCDVIN, RCONDE, RCNDE1, RCDEIN,
00928      $             SCALE, SCALE1, RESULT, WORK, 6*N+2*N**2, IWORK,
00929      $             INFO )
00930 *
00931 *     Check for RESULT(j) > THRESH
00932 *
00933       NTEST = 0
00934       NFAIL = 0
00935       DO 200 J = 1, 11
00936          IF( RESULT( J ).GE.ZERO )
00937      $      NTEST = NTEST + 1
00938          IF( RESULT( J ).GE.THRESH )
00939      $      NFAIL = NFAIL + 1
00940   200 CONTINUE
00941 *
00942       IF( NFAIL.GT.0 )
00943      $   NTESTF = NTESTF + 1
00944       IF( NTESTF.EQ.1 ) THEN
00945          WRITE( NOUNIT, FMT = 9999 )PATH
00946          WRITE( NOUNIT, FMT = 9998 )
00947          WRITE( NOUNIT, FMT = 9997 )
00948          WRITE( NOUNIT, FMT = 9996 )
00949          WRITE( NOUNIT, FMT = 9995 )THRESH
00950          NTESTF = 2
00951       END IF
00952 *
00953       DO 210 J = 1, 11
00954          IF( RESULT( J ).GE.THRESH ) THEN
00955             WRITE( NOUNIT, FMT = 9993 )N, JTYPE, J, RESULT( J )
00956          END IF
00957   210 CONTINUE
00958 *
00959       NERRS = NERRS + NFAIL
00960       NTESTT = NTESTT + NTEST
00961       GO TO 170
00962   220 CONTINUE
00963 *
00964 *     Summary
00965 *
00966       CALL SLASUM( PATH, NOUNIT, NERRS, NTESTT )
00967 *
00968  9999 FORMAT( / 1X, A3, ' -- Real Eigenvalue-Eigenvector Decomposition',
00969      $      ' Expert Driver', /
00970      $      ' Matrix types (see SDRVVX for details): ' )
00971 *
00972  9998 FORMAT( / ' Special Matrices:', / '  1=Zero matrix.             ',
00973      $      '           ', '  5=Diagonal: geometr. spaced entries.',
00974      $      / '  2=Identity matrix.                    ', '  6=Diagona',
00975      $      'l: clustered entries.', / '  3=Transposed Jordan block.  ',
00976      $      '          ', '  7=Diagonal: large, evenly spaced.', / '  ',
00977      $      '4=Diagonal: evenly spaced entries.    ', '  8=Diagonal: s',
00978      $      'mall, evenly spaced.' )
00979  9997 FORMAT( ' Dense, Non-Symmetric Matrices:', / '  9=Well-cond., ev',
00980      $      'enly spaced eigenvals.', ' 14=Ill-cond., geomet. spaced e',
00981      $      'igenals.', / ' 10=Well-cond., geom. spaced eigenvals. ',
00982      $      ' 15=Ill-conditioned, clustered e.vals.', / ' 11=Well-cond',
00983      $      'itioned, clustered e.vals. ', ' 16=Ill-cond., random comp',
00984      $      'lex ', / ' 12=Well-cond., random complex ', '         ',
00985      $      ' 17=Ill-cond., large rand. complx ', / ' 13=Ill-condi',
00986      $      'tioned, evenly spaced.     ', ' 18=Ill-cond., small rand.',
00987      $      ' complx ' )
00988  9996 FORMAT( ' 19=Matrix with random O(1) entries.    ', ' 21=Matrix ',
00989      $      'with small random entries.', / ' 20=Matrix with large ran',
00990      $      'dom entries.   ', ' 22=Matrix read from input file', / )
00991  9995 FORMAT( ' Tests performed with test threshold =', F8.2,
00992      $      / / ' 1 = | A VR - VR W | / ( n |A| ulp ) ',
00993      $      / ' 2 = | transpose(A) VL - VL W | / ( n |A| ulp ) ',
00994      $      / ' 3 = | |VR(i)| - 1 | / ulp ',
00995      $      / ' 4 = | |VL(i)| - 1 | / ulp ',
00996      $      / ' 5 = 0 if W same no matter if VR or VL computed,',
00997      $      ' 1/ulp otherwise', /
00998      $      ' 6 = 0 if VR same no matter what else computed,',
00999      $      '  1/ulp otherwise', /
01000      $      ' 7 = 0 if VL same no matter what else computed,',
01001      $      '  1/ulp otherwise', /
01002      $      ' 8 = 0 if RCONDV same no matter what else computed,',
01003      $      '  1/ulp otherwise', /
01004      $      ' 9 = 0 if SCALE, ILO, IHI, ABNRM same no matter what else',
01005      $      ' computed,  1/ulp otherwise',
01006      $      / ' 10 = | RCONDV - RCONDV(precomputed) | / cond(RCONDV),',
01007      $      / ' 11 = | RCONDE - RCONDE(precomputed) | / cond(RCONDE),' )
01008  9994 FORMAT( ' BALANC=''', A1, ''',N=', I4, ',IWK=', I1, ', seed=',
01009      $      4( I4, ',' ), ' type ', I2, ', test(', I2, ')=', G10.3 )
01010  9993 FORMAT( ' N=', I5, ', input example =', I3, ',  test(', I2, ')=',
01011      $      G10.3 )
01012  9992 FORMAT( ' SDRVVX: ', A, ' returned INFO=', I6, '.', / 9X, 'N=',
01013      $      I6, ', JTYPE=', I6, ', ISEED=(', 3( I5, ',' ), I5, ')' )
01014 *
01015       RETURN
01016 *
01017 *     End of SDRVVX
01018 *
01019       END
 All Files Functions