File:  [local] / rpl / lapack / lapack / dpbstf.f
Revision 1.11: download - view: text, annotated - select for diffs - revision graph
Fri Dec 14 14:22:38 2012 UTC (11 years, 5 months ago) by bertrand
Branches: MAIN
CVS tags: rpl-4_1_16, rpl-4_1_15, rpl-4_1_14, rpl-4_1_13, rpl-4_1_12, rpl-4_1_11, HEAD
Mise à jour de lapack.

    1: *> \brief \b DPBSTF
    2: *
    3: *  =========== DOCUMENTATION ===========
    4: *
    5: * Online html documentation available at 
    6: *            http://www.netlib.org/lapack/explore-html/ 
    7: *
    8: *> \htmlonly
    9: *> Download DPBSTF + dependencies 
   10: *> <a href="http://www.netlib.org/cgi-bin/netlibfiles.tgz?format=tgz&filename=/lapack/lapack_routine/dpbstf.f"> 
   11: *> [TGZ]</a> 
   12: *> <a href="http://www.netlib.org/cgi-bin/netlibfiles.zip?format=zip&filename=/lapack/lapack_routine/dpbstf.f"> 
   13: *> [ZIP]</a> 
   14: *> <a href="http://www.netlib.org/cgi-bin/netlibfiles.txt?format=txt&filename=/lapack/lapack_routine/dpbstf.f"> 
   15: *> [TXT]</a>
   16: *> \endhtmlonly 
   17: *
   18: *  Definition:
   19: *  ===========
   20: *
   21: *       SUBROUTINE DPBSTF( UPLO, N, KD, AB, LDAB, INFO )
   22:    23: *       .. Scalar Arguments ..
   24: *       CHARACTER          UPLO
   25: *       INTEGER            INFO, KD, LDAB, N
   26: *       ..
   27: *       .. Array Arguments ..
   28: *       DOUBLE PRECISION   AB( LDAB, * )
   29: *       ..
   30: *  
   31: *
   32: *> \par Purpose:
   33: *  =============
   34: *>
   35: *> \verbatim
   36: *>
   37: *> DPBSTF computes a split Cholesky factorization of a real
   38: *> symmetric positive definite band matrix A.
   39: *>
   40: *> This routine is designed to be used in conjunction with DSBGST.
   41: *>
   42: *> The factorization has the form  A = S**T*S  where S is a band matrix
   43: *> of the same bandwidth as A and the following structure:
   44: *>
   45: *>   S = ( U    )
   46: *>       ( M  L )
   47: *>
   48: *> where U is upper triangular of order m = (n+kd)/2, and L is lower
   49: *> triangular of order n-m.
   50: *> \endverbatim
   51: *
   52: *  Arguments:
   53: *  ==========
   54: *
   55: *> \param[in] UPLO
   56: *> \verbatim
   57: *>          UPLO is CHARACTER*1
   58: *>          = 'U':  Upper triangle of A is stored;
   59: *>          = 'L':  Lower triangle of A is stored.
   60: *> \endverbatim
   61: *>
   62: *> \param[in] N
   63: *> \verbatim
   64: *>          N is INTEGER
   65: *>          The order of the matrix A.  N >= 0.
   66: *> \endverbatim
   67: *>
   68: *> \param[in] KD
   69: *> \verbatim
   70: *>          KD is INTEGER
   71: *>          The number of superdiagonals of the matrix A if UPLO = 'U',
   72: *>          or the number of subdiagonals if UPLO = 'L'.  KD >= 0.
   73: *> \endverbatim
   74: *>
   75: *> \param[in,out] AB
   76: *> \verbatim
   77: *>          AB is DOUBLE PRECISION array, dimension (LDAB,N)
   78: *>          On entry, the upper or lower triangle of the symmetric band
   79: *>          matrix A, stored in the first kd+1 rows of the array.  The
   80: *>          j-th column of A is stored in the j-th column of the array AB
   81: *>          as follows:
   82: *>          if UPLO = 'U', AB(kd+1+i-j,j) = A(i,j) for max(1,j-kd)<=i<=j;
   83: *>          if UPLO = 'L', AB(1+i-j,j)    = A(i,j) for j<=i<=min(n,j+kd).
   84: *>
   85: *>          On exit, if INFO = 0, the factor S from the split Cholesky
   86: *>          factorization A = S**T*S. See Further Details.
   87: *> \endverbatim
   88: *>
   89: *> \param[in] LDAB
   90: *> \verbatim
   91: *>          LDAB is INTEGER
   92: *>          The leading dimension of the array AB.  LDAB >= KD+1.
   93: *> \endverbatim
   94: *>
   95: *> \param[out] INFO
   96: *> \verbatim
   97: *>          INFO is INTEGER
   98: *>          = 0: successful exit
   99: *>          < 0: if INFO = -i, the i-th argument had an illegal value
  100: *>          > 0: if INFO = i, the factorization could not be completed,
  101: *>               because the updated element a(i,i) was negative; the
  102: *>               matrix A is not positive definite.
  103: *> \endverbatim
  104: *
  105: *  Authors:
  106: *  ========
  107: *
  108: *> \author Univ. of Tennessee 
  109: *> \author Univ. of California Berkeley 
  110: *> \author Univ. of Colorado Denver 
  111: *> \author NAG Ltd. 
  112: *
  113: *> \date November 2011
  114: *
  115: *> \ingroup doubleOTHERcomputational
  116: *
  117: *> \par Further Details:
  118: *  =====================
  119: *>
  120: *> \verbatim
  121: *>
  122: *>  The band storage scheme is illustrated by the following example, when
  123: *>  N = 7, KD = 2:
  124: *>
  125: *>  S = ( s11  s12  s13                     )
  126: *>      (      s22  s23  s24                )
  127: *>      (           s33  s34                )
  128: *>      (                s44                )
  129: *>      (           s53  s54  s55           )
  130: *>      (                s64  s65  s66      )
  131: *>      (                     s75  s76  s77 )
  132: *>
  133: *>  If UPLO = 'U', the array AB holds:
  134: *>
  135: *>  on entry:                          on exit:
  136: *>
  137: *>   *    *   a13  a24  a35  a46  a57   *    *   s13  s24  s53  s64  s75
  138: *>   *   a12  a23  a34  a45  a56  a67   *   s12  s23  s34  s54  s65  s76
  139: *>  a11  a22  a33  a44  a55  a66  a77  s11  s22  s33  s44  s55  s66  s77
  140: *>
  141: *>  If UPLO = 'L', the array AB holds:
  142: *>
  143: *>  on entry:                          on exit:
  144: *>
  145: *>  a11  a22  a33  a44  a55  a66  a77  s11  s22  s33  s44  s55  s66  s77
  146: *>  a21  a32  a43  a54  a65  a76   *   s12  s23  s34  s54  s65  s76   *
  147: *>  a31  a42  a53  a64  a64   *    *   s13  s24  s53  s64  s75   *    *
  148: *>
  149: *>  Array elements marked * are not used by the routine.
  150: *> \endverbatim
  151: *>
  152: *  =====================================================================
  153:       SUBROUTINE DPBSTF( UPLO, N, KD, AB, LDAB, INFO )
  154: *
  155: *  -- LAPACK computational routine (version 3.4.0) --
  156: *  -- LAPACK is a software package provided by Univ. of Tennessee,    --
  157: *  -- Univ. of California Berkeley, Univ. of Colorado Denver and NAG Ltd..--
  158: *     November 2011
  159: *
  160: *     .. Scalar Arguments ..
  161:       CHARACTER          UPLO
  162:       INTEGER            INFO, KD, LDAB, N
  163: *     ..
  164: *     .. Array Arguments ..
  165:       DOUBLE PRECISION   AB( LDAB, * )
  166: *     ..
  167: *
  168: *  =====================================================================
  169: *
  170: *     .. Parameters ..
  171:       DOUBLE PRECISION   ONE, ZERO
  172:       PARAMETER          ( ONE = 1.0D+0, ZERO = 0.0D+0 )
  173: *     ..
  174: *     .. Local Scalars ..
  175:       LOGICAL            UPPER
  176:       INTEGER            J, KLD, KM, M
  177:       DOUBLE PRECISION   AJJ
  178: *     ..
  179: *     .. External Functions ..
  180:       LOGICAL            LSAME
  181:       EXTERNAL           LSAME
  182: *     ..
  183: *     .. External Subroutines ..
  184:       EXTERNAL           DSCAL, DSYR, XERBLA
  185: *     ..
  186: *     .. Intrinsic Functions ..
  187:       INTRINSIC          MAX, MIN, SQRT
  188: *     ..
  189: *     .. Executable Statements ..
  190: *
  191: *     Test the input parameters.
  192: *
  193:       INFO = 0
  194:       UPPER = LSAME( UPLO, 'U' )
  195:       IF( .NOT.UPPER .AND. .NOT.LSAME( UPLO, 'L' ) ) THEN
  196:          INFO = -1
  197:       ELSE IF( N.LT.0 ) THEN
  198:          INFO = -2
  199:       ELSE IF( KD.LT.0 ) THEN
  200:          INFO = -3
  201:       ELSE IF( LDAB.LT.KD+1 ) THEN
  202:          INFO = -5
  203:       END IF
  204:       IF( INFO.NE.0 ) THEN
  205:          CALL XERBLA( 'DPBSTF', -INFO )
  206:          RETURN
  207:       END IF
  208: *
  209: *     Quick return if possible
  210: *
  211:       IF( N.EQ.0 )
  212:      $   RETURN
  213: *
  214:       KLD = MAX( 1, LDAB-1 )
  215: *
  216: *     Set the splitting point m.
  217: *
  218:       M = ( N+KD ) / 2
  219: *
  220:       IF( UPPER ) THEN
  221: *
  222: *        Factorize A(m+1:n,m+1:n) as L**T*L, and update A(1:m,1:m).
  223: *
  224:          DO 10 J = N, M + 1, -1
  225: *
  226: *           Compute s(j,j) and test for non-positive-definiteness.
  227: *
  228:             AJJ = AB( KD+1, J )
  229:             IF( AJJ.LE.ZERO )
  230:      $         GO TO 50
  231:             AJJ = SQRT( AJJ )
  232:             AB( KD+1, J ) = AJJ
  233:             KM = MIN( J-1, KD )
  234: *
  235: *           Compute elements j-km:j-1 of the j-th column and update the
  236: *           the leading submatrix within the band.
  237: *
  238:             CALL DSCAL( KM, ONE / AJJ, AB( KD+1-KM, J ), 1 )
  239:             CALL DSYR( 'Upper', KM, -ONE, AB( KD+1-KM, J ), 1,
  240:      $                 AB( KD+1, J-KM ), KLD )
  241:    10    CONTINUE
  242: *
  243: *        Factorize the updated submatrix A(1:m,1:m) as U**T*U.
  244: *
  245:          DO 20 J = 1, M
  246: *
  247: *           Compute s(j,j) and test for non-positive-definiteness.
  248: *
  249:             AJJ = AB( KD+1, J )
  250:             IF( AJJ.LE.ZERO )
  251:      $         GO TO 50
  252:             AJJ = SQRT( AJJ )
  253:             AB( KD+1, J ) = AJJ
  254:             KM = MIN( KD, M-J )
  255: *
  256: *           Compute elements j+1:j+km of the j-th row and update the
  257: *           trailing submatrix within the band.
  258: *
  259:             IF( KM.GT.0 ) THEN
  260:                CALL DSCAL( KM, ONE / AJJ, AB( KD, J+1 ), KLD )
  261:                CALL DSYR( 'Upper', KM, -ONE, AB( KD, J+1 ), KLD,
  262:      $                    AB( KD+1, J+1 ), KLD )
  263:             END IF
  264:    20    CONTINUE
  265:       ELSE
  266: *
  267: *        Factorize A(m+1:n,m+1:n) as L**T*L, and update A(1:m,1:m).
  268: *
  269:          DO 30 J = N, M + 1, -1
  270: *
  271: *           Compute s(j,j) and test for non-positive-definiteness.
  272: *
  273:             AJJ = AB( 1, J )
  274:             IF( AJJ.LE.ZERO )
  275:      $         GO TO 50
  276:             AJJ = SQRT( AJJ )
  277:             AB( 1, J ) = AJJ
  278:             KM = MIN( J-1, KD )
  279: *
  280: *           Compute elements j-km:j-1 of the j-th row and update the
  281: *           trailing submatrix within the band.
  282: *
  283:             CALL DSCAL( KM, ONE / AJJ, AB( KM+1, J-KM ), KLD )
  284:             CALL DSYR( 'Lower', KM, -ONE, AB( KM+1, J-KM ), KLD,
  285:      $                 AB( 1, J-KM ), KLD )
  286:    30    CONTINUE
  287: *
  288: *        Factorize the updated submatrix A(1:m,1:m) as U**T*U.
  289: *
  290:          DO 40 J = 1, M
  291: *
  292: *           Compute s(j,j) and test for non-positive-definiteness.
  293: *
  294:             AJJ = AB( 1, J )
  295:             IF( AJJ.LE.ZERO )
  296:      $         GO TO 50
  297:             AJJ = SQRT( AJJ )
  298:             AB( 1, J ) = AJJ
  299:             KM = MIN( KD, M-J )
  300: *
  301: *           Compute elements j+1:j+km of the j-th column and update the
  302: *           trailing submatrix within the band.
  303: *
  304:             IF( KM.GT.0 ) THEN
  305:                CALL DSCAL( KM, ONE / AJJ, AB( 2, J ), 1 )
  306:                CALL DSYR( 'Lower', KM, -ONE, AB( 2, J ), 1,
  307:      $                    AB( 1, J+1 ), KLD )
  308:             END IF
  309:    40    CONTINUE
  310:       END IF
  311:       RETURN
  312: *
  313:    50 CONTINUE
  314:       INFO = J
  315:       RETURN
  316: *
  317: *     End of DPBSTF
  318: *
  319:       END

CVSweb interface <joel.bertrand@systella.fr>