clals0.c
Go to the documentation of this file.
00001 /* clals0.f -- translated by f2c (version 20061008).
00002    You must link the resulting object file with libf2c:
00003         on Microsoft Windows system, link with libf2c.lib;
00004         on Linux or Unix systems, link with .../path/to/libf2c.a -lm
00005         or, if you install libf2c.a in a standard place, with -lf2c -lm
00006         -- in that order, at the end of the command line, as in
00007                 cc *.o -lf2c -lm
00008         Source for libf2c is in /netlib/f2c/libf2c.zip, e.g.,
00009 
00010                 http://www.netlib.org/f2c/libf2c.zip
00011 */
00012 
00013 #include "f2c.h"
00014 #include "blaswrap.h"
00015 
00016 /* Table of constant values */
00017 
00018 static real c_b5 = -1.f;
00019 static integer c__1 = 1;
00020 static real c_b13 = 1.f;
00021 static real c_b15 = 0.f;
00022 static integer c__0 = 0;
00023 
00024 /* Subroutine */ int clals0_(integer *icompq, integer *nl, integer *nr, 
00025         integer *sqre, integer *nrhs, complex *b, integer *ldb, complex *bx, 
00026         integer *ldbx, integer *perm, integer *givptr, integer *givcol, 
00027         integer *ldgcol, real *givnum, integer *ldgnum, real *poles, real *
00028         difl, real *difr, real *z__, integer *k, real *c__, real *s, real *
00029         rwork, integer *info)
00030 {
00031     /* System generated locals */
00032     integer givcol_dim1, givcol_offset, difr_dim1, difr_offset, givnum_dim1, 
00033             givnum_offset, poles_dim1, poles_offset, b_dim1, b_offset, 
00034             bx_dim1, bx_offset, i__1, i__2, i__3, i__4, i__5;
00035     real r__1;
00036     complex q__1;
00037 
00038     /* Builtin functions */
00039     double r_imag(complex *);
00040 
00041     /* Local variables */
00042     integer i__, j, m, n;
00043     real dj;
00044     integer nlp1, jcol;
00045     real temp;
00046     integer jrow;
00047     extern doublereal snrm2_(integer *, real *, integer *);
00048     real diflj, difrj, dsigj;
00049     extern /* Subroutine */ int ccopy_(integer *, complex *, integer *, 
00050             complex *, integer *), sgemv_(char *, integer *, integer *, real *
00051 , real *, integer *, real *, integer *, real *, real *, integer *), csrot_(integer *, complex *, integer *, complex *, 
00052             integer *, real *, real *);
00053     extern doublereal slamc3_(real *, real *);
00054     extern /* Subroutine */ int clascl_(char *, integer *, integer *, real *, 
00055             real *, integer *, integer *, complex *, integer *, integer *), csscal_(integer *, real *, complex *, integer *), 
00056             clacpy_(char *, integer *, integer *, complex *, integer *, 
00057             complex *, integer *), xerbla_(char *, integer *);
00058     real dsigjp;
00059 
00060 
00061 /*  -- LAPACK routine (version 3.2) -- */
00062 /*     Univ. of Tennessee, Univ. of California Berkeley and NAG Ltd.. */
00063 /*     November 2006 */
00064 
00065 /*     .. Scalar Arguments .. */
00066 /*     .. */
00067 /*     .. Array Arguments .. */
00068 /*     .. */
00069 
00070 /*  Purpose */
00071 /*  ======= */
00072 
00073 /*  CLALS0 applies back the multiplying factors of either the left or the */
00074 /*  right singular vector matrix of a diagonal matrix appended by a row */
00075 /*  to the right hand side matrix B in solving the least squares problem */
00076 /*  using the divide-and-conquer SVD approach. */
00077 
00078 /*  For the left singular vector matrix, three types of orthogonal */
00079 /*  matrices are involved: */
00080 
00081 /*  (1L) Givens rotations: the number of such rotations is GIVPTR; the */
00082 /*       pairs of columns/rows they were applied to are stored in GIVCOL; */
00083 /*       and the C- and S-values of these rotations are stored in GIVNUM. */
00084 
00085 /*  (2L) Permutation. The (NL+1)-st row of B is to be moved to the first */
00086 /*       row, and for J=2:N, PERM(J)-th row of B is to be moved to the */
00087 /*       J-th row. */
00088 
00089 /*  (3L) The left singular vector matrix of the remaining matrix. */
00090 
00091 /*  For the right singular vector matrix, four types of orthogonal */
00092 /*  matrices are involved: */
00093 
00094 /*  (1R) The right singular vector matrix of the remaining matrix. */
00095 
00096 /*  (2R) If SQRE = 1, one extra Givens rotation to generate the right */
00097 /*       null space. */
00098 
00099 /*  (3R) The inverse transformation of (2L). */
00100 
00101 /*  (4R) The inverse transformation of (1L). */
00102 
00103 /*  Arguments */
00104 /*  ========= */
00105 
00106 /*  ICOMPQ (input) INTEGER */
00107 /*         Specifies whether singular vectors are to be computed in */
00108 /*         factored form: */
00109 /*         = 0: Left singular vector matrix. */
00110 /*         = 1: Right singular vector matrix. */
00111 
00112 /*  NL     (input) INTEGER */
00113 /*         The row dimension of the upper block. NL >= 1. */
00114 
00115 /*  NR     (input) INTEGER */
00116 /*         The row dimension of the lower block. NR >= 1. */
00117 
00118 /*  SQRE   (input) INTEGER */
00119 /*         = 0: the lower block is an NR-by-NR square matrix. */
00120 /*         = 1: the lower block is an NR-by-(NR+1) rectangular matrix. */
00121 
00122 /*         The bidiagonal matrix has row dimension N = NL + NR + 1, */
00123 /*         and column dimension M = N + SQRE. */
00124 
00125 /*  NRHS   (input) INTEGER */
00126 /*         The number of columns of B and BX. NRHS must be at least 1. */
00127 
00128 /*  B      (input/output) COMPLEX array, dimension ( LDB, NRHS ) */
00129 /*         On input, B contains the right hand sides of the least */
00130 /*         squares problem in rows 1 through M. On output, B contains */
00131 /*         the solution X in rows 1 through N. */
00132 
00133 /*  LDB    (input) INTEGER */
00134 /*         The leading dimension of B. LDB must be at least */
00135 /*         max(1,MAX( M, N ) ). */
00136 
00137 /*  BX     (workspace) COMPLEX array, dimension ( LDBX, NRHS ) */
00138 
00139 /*  LDBX   (input) INTEGER */
00140 /*         The leading dimension of BX. */
00141 
00142 /*  PERM   (input) INTEGER array, dimension ( N ) */
00143 /*         The permutations (from deflation and sorting) applied */
00144 /*         to the two blocks. */
00145 
00146 /*  GIVPTR (input) INTEGER */
00147 /*         The number of Givens rotations which took place in this */
00148 /*         subproblem. */
00149 
00150 /*  GIVCOL (input) INTEGER array, dimension ( LDGCOL, 2 ) */
00151 /*         Each pair of numbers indicates a pair of rows/columns */
00152 /*         involved in a Givens rotation. */
00153 
00154 /*  LDGCOL (input) INTEGER */
00155 /*         The leading dimension of GIVCOL, must be at least N. */
00156 
00157 /*  GIVNUM (input) REAL array, dimension ( LDGNUM, 2 ) */
00158 /*         Each number indicates the C or S value used in the */
00159 /*         corresponding Givens rotation. */
00160 
00161 /*  LDGNUM (input) INTEGER */
00162 /*         The leading dimension of arrays DIFR, POLES and */
00163 /*         GIVNUM, must be at least K. */
00164 
00165 /*  POLES  (input) REAL array, dimension ( LDGNUM, 2 ) */
00166 /*         On entry, POLES(1:K, 1) contains the new singular */
00167 /*         values obtained from solving the secular equation, and */
00168 /*         POLES(1:K, 2) is an array containing the poles in the secular */
00169 /*         equation. */
00170 
00171 /*  DIFL   (input) REAL array, dimension ( K ). */
00172 /*         On entry, DIFL(I) is the distance between I-th updated */
00173 /*         (undeflated) singular value and the I-th (undeflated) old */
00174 /*         singular value. */
00175 
00176 /*  DIFR   (input) REAL array, dimension ( LDGNUM, 2 ). */
00177 /*         On entry, DIFR(I, 1) contains the distances between I-th */
00178 /*         updated (undeflated) singular value and the I+1-th */
00179 /*         (undeflated) old singular value. And DIFR(I, 2) is the */
00180 /*         normalizing factor for the I-th right singular vector. */
00181 
00182 /*  Z      (input) REAL array, dimension ( K ) */
00183 /*         Contain the components of the deflation-adjusted updating row */
00184 /*         vector. */
00185 
00186 /*  K      (input) INTEGER */
00187 /*         Contains the dimension of the non-deflated matrix, */
00188 /*         This is the order of the related secular equation. 1 <= K <=N. */
00189 
00190 /*  C      (input) REAL */
00191 /*         C contains garbage if SQRE =0 and the C-value of a Givens */
00192 /*         rotation related to the right null space if SQRE = 1. */
00193 
00194 /*  S      (input) REAL */
00195 /*         S contains garbage if SQRE =0 and the S-value of a Givens */
00196 /*         rotation related to the right null space if SQRE = 1. */
00197 
00198 /*  RWORK  (workspace) REAL array, dimension */
00199 /*         ( K*(1+NRHS) + 2*NRHS ) */
00200 
00201 /*  INFO   (output) INTEGER */
00202 /*          = 0:  successful exit. */
00203 /*          < 0:  if INFO = -i, the i-th argument had an illegal value. */
00204 
00205 /*  Further Details */
00206 /*  =============== */
00207 
00208 /*  Based on contributions by */
00209 /*     Ming Gu and Ren-Cang Li, Computer Science Division, University of */
00210 /*       California at Berkeley, USA */
00211 /*     Osni Marques, LBNL/NERSC, USA */
00212 
00213 /*  ===================================================================== */
00214 
00215 /*     .. Parameters .. */
00216 /*     .. */
00217 /*     .. Local Scalars .. */
00218 /*     .. */
00219 /*     .. External Subroutines .. */
00220 /*     .. */
00221 /*     .. External Functions .. */
00222 /*     .. */
00223 /*     .. Intrinsic Functions .. */
00224 /*     .. */
00225 /*     .. Executable Statements .. */
00226 
00227 /*     Test the input parameters. */
00228 
00229     /* Parameter adjustments */
00230     b_dim1 = *ldb;
00231     b_offset = 1 + b_dim1;
00232     b -= b_offset;
00233     bx_dim1 = *ldbx;
00234     bx_offset = 1 + bx_dim1;
00235     bx -= bx_offset;
00236     --perm;
00237     givcol_dim1 = *ldgcol;
00238     givcol_offset = 1 + givcol_dim1;
00239     givcol -= givcol_offset;
00240     difr_dim1 = *ldgnum;
00241     difr_offset = 1 + difr_dim1;
00242     difr -= difr_offset;
00243     poles_dim1 = *ldgnum;
00244     poles_offset = 1 + poles_dim1;
00245     poles -= poles_offset;
00246     givnum_dim1 = *ldgnum;
00247     givnum_offset = 1 + givnum_dim1;
00248     givnum -= givnum_offset;
00249     --difl;
00250     --z__;
00251     --rwork;
00252 
00253     /* Function Body */
00254     *info = 0;
00255 
00256     if (*icompq < 0 || *icompq > 1) {
00257         *info = -1;
00258     } else if (*nl < 1) {
00259         *info = -2;
00260     } else if (*nr < 1) {
00261         *info = -3;
00262     } else if (*sqre < 0 || *sqre > 1) {
00263         *info = -4;
00264     }
00265 
00266     n = *nl + *nr + 1;
00267 
00268     if (*nrhs < 1) {
00269         *info = -5;
00270     } else if (*ldb < n) {
00271         *info = -7;
00272     } else if (*ldbx < n) {
00273         *info = -9;
00274     } else if (*givptr < 0) {
00275         *info = -11;
00276     } else if (*ldgcol < n) {
00277         *info = -13;
00278     } else if (*ldgnum < n) {
00279         *info = -15;
00280     } else if (*k < 1) {
00281         *info = -20;
00282     }
00283     if (*info != 0) {
00284         i__1 = -(*info);
00285         xerbla_("CLALS0", &i__1);
00286         return 0;
00287     }
00288 
00289     m = n + *sqre;
00290     nlp1 = *nl + 1;
00291 
00292     if (*icompq == 0) {
00293 
00294 /*        Apply back orthogonal transformations from the left. */
00295 
00296 /*        Step (1L): apply back the Givens rotations performed. */
00297 
00298         i__1 = *givptr;
00299         for (i__ = 1; i__ <= i__1; ++i__) {
00300             csrot_(nrhs, &b[givcol[i__ + (givcol_dim1 << 1)] + b_dim1], ldb, &
00301                     b[givcol[i__ + givcol_dim1] + b_dim1], ldb, &givnum[i__ + 
00302                     (givnum_dim1 << 1)], &givnum[i__ + givnum_dim1]);
00303 /* L10: */
00304         }
00305 
00306 /*        Step (2L): permute rows of B. */
00307 
00308         ccopy_(nrhs, &b[nlp1 + b_dim1], ldb, &bx[bx_dim1 + 1], ldbx);
00309         i__1 = n;
00310         for (i__ = 2; i__ <= i__1; ++i__) {
00311             ccopy_(nrhs, &b[perm[i__] + b_dim1], ldb, &bx[i__ + bx_dim1], 
00312                     ldbx);
00313 /* L20: */
00314         }
00315 
00316 /*        Step (3L): apply the inverse of the left singular vector */
00317 /*        matrix to BX. */
00318 
00319         if (*k == 1) {
00320             ccopy_(nrhs, &bx[bx_offset], ldbx, &b[b_offset], ldb);
00321             if (z__[1] < 0.f) {
00322                 csscal_(nrhs, &c_b5, &b[b_offset], ldb);
00323             }
00324         } else {
00325             i__1 = *k;
00326             for (j = 1; j <= i__1; ++j) {
00327                 diflj = difl[j];
00328                 dj = poles[j + poles_dim1];
00329                 dsigj = -poles[j + (poles_dim1 << 1)];
00330                 if (j < *k) {
00331                     difrj = -difr[j + difr_dim1];
00332                     dsigjp = -poles[j + 1 + (poles_dim1 << 1)];
00333                 }
00334                 if (z__[j] == 0.f || poles[j + (poles_dim1 << 1)] == 0.f) {
00335                     rwork[j] = 0.f;
00336                 } else {
00337                     rwork[j] = -poles[j + (poles_dim1 << 1)] * z__[j] / diflj 
00338                             / (poles[j + (poles_dim1 << 1)] + dj);
00339                 }
00340                 i__2 = j - 1;
00341                 for (i__ = 1; i__ <= i__2; ++i__) {
00342                     if (z__[i__] == 0.f || poles[i__ + (poles_dim1 << 1)] == 
00343                             0.f) {
00344                         rwork[i__] = 0.f;
00345                     } else {
00346                         rwork[i__] = poles[i__ + (poles_dim1 << 1)] * z__[i__]
00347                                  / (slamc3_(&poles[i__ + (poles_dim1 << 1)], &
00348                                 dsigj) - diflj) / (poles[i__ + (poles_dim1 << 
00349                                 1)] + dj);
00350                     }
00351 /* L30: */
00352                 }
00353                 i__2 = *k;
00354                 for (i__ = j + 1; i__ <= i__2; ++i__) {
00355                     if (z__[i__] == 0.f || poles[i__ + (poles_dim1 << 1)] == 
00356                             0.f) {
00357                         rwork[i__] = 0.f;
00358                     } else {
00359                         rwork[i__] = poles[i__ + (poles_dim1 << 1)] * z__[i__]
00360                                  / (slamc3_(&poles[i__ + (poles_dim1 << 1)], &
00361                                 dsigjp) + difrj) / (poles[i__ + (poles_dim1 <<
00362                                  1)] + dj);
00363                     }
00364 /* L40: */
00365                 }
00366                 rwork[1] = -1.f;
00367                 temp = snrm2_(k, &rwork[1], &c__1);
00368 
00369 /*              Since B and BX are complex, the following call to SGEMV */
00370 /*              is performed in two steps (real and imaginary parts). */
00371 
00372 /*              CALL SGEMV( 'T', K, NRHS, ONE, BX, LDBX, WORK, 1, ZERO, */
00373 /*    $                     B( J, 1 ), LDB ) */
00374 
00375                 i__ = *k + (*nrhs << 1);
00376                 i__2 = *nrhs;
00377                 for (jcol = 1; jcol <= i__2; ++jcol) {
00378                     i__3 = *k;
00379                     for (jrow = 1; jrow <= i__3; ++jrow) {
00380                         ++i__;
00381                         i__4 = jrow + jcol * bx_dim1;
00382                         rwork[i__] = bx[i__4].r;
00383 /* L50: */
00384                     }
00385 /* L60: */
00386                 }
00387                 sgemv_("T", k, nrhs, &c_b13, &rwork[*k + 1 + (*nrhs << 1)], k, 
00388                          &rwork[1], &c__1, &c_b15, &rwork[*k + 1], &c__1);
00389                 i__ = *k + (*nrhs << 1);
00390                 i__2 = *nrhs;
00391                 for (jcol = 1; jcol <= i__2; ++jcol) {
00392                     i__3 = *k;
00393                     for (jrow = 1; jrow <= i__3; ++jrow) {
00394                         ++i__;
00395                         rwork[i__] = r_imag(&bx[jrow + jcol * bx_dim1]);
00396 /* L70: */
00397                     }
00398 /* L80: */
00399                 }
00400                 sgemv_("T", k, nrhs, &c_b13, &rwork[*k + 1 + (*nrhs << 1)], k, 
00401                          &rwork[1], &c__1, &c_b15, &rwork[*k + 1 + *nrhs], &
00402                         c__1);
00403                 i__2 = *nrhs;
00404                 for (jcol = 1; jcol <= i__2; ++jcol) {
00405                     i__3 = j + jcol * b_dim1;
00406                     i__4 = jcol + *k;
00407                     i__5 = jcol + *k + *nrhs;
00408                     q__1.r = rwork[i__4], q__1.i = rwork[i__5];
00409                     b[i__3].r = q__1.r, b[i__3].i = q__1.i;
00410 /* L90: */
00411                 }
00412                 clascl_("G", &c__0, &c__0, &temp, &c_b13, &c__1, nrhs, &b[j + 
00413                         b_dim1], ldb, info);
00414 /* L100: */
00415             }
00416         }
00417 
00418 /*        Move the deflated rows of BX to B also. */
00419 
00420         if (*k < max(m,n)) {
00421             i__1 = n - *k;
00422             clacpy_("A", &i__1, nrhs, &bx[*k + 1 + bx_dim1], ldbx, &b[*k + 1 
00423                     + b_dim1], ldb);
00424         }
00425     } else {
00426 
00427 /*        Apply back the right orthogonal transformations. */
00428 
00429 /*        Step (1R): apply back the new right singular vector matrix */
00430 /*        to B. */
00431 
00432         if (*k == 1) {
00433             ccopy_(nrhs, &b[b_offset], ldb, &bx[bx_offset], ldbx);
00434         } else {
00435             i__1 = *k;
00436             for (j = 1; j <= i__1; ++j) {
00437                 dsigj = poles[j + (poles_dim1 << 1)];
00438                 if (z__[j] == 0.f) {
00439                     rwork[j] = 0.f;
00440                 } else {
00441                     rwork[j] = -z__[j] / difl[j] / (dsigj + poles[j + 
00442                             poles_dim1]) / difr[j + (difr_dim1 << 1)];
00443                 }
00444                 i__2 = j - 1;
00445                 for (i__ = 1; i__ <= i__2; ++i__) {
00446                     if (z__[j] == 0.f) {
00447                         rwork[i__] = 0.f;
00448                     } else {
00449                         r__1 = -poles[i__ + 1 + (poles_dim1 << 1)];
00450                         rwork[i__] = z__[j] / (slamc3_(&dsigj, &r__1) - difr[
00451                                 i__ + difr_dim1]) / (dsigj + poles[i__ + 
00452                                 poles_dim1]) / difr[i__ + (difr_dim1 << 1)];
00453                     }
00454 /* L110: */
00455                 }
00456                 i__2 = *k;
00457                 for (i__ = j + 1; i__ <= i__2; ++i__) {
00458                     if (z__[j] == 0.f) {
00459                         rwork[i__] = 0.f;
00460                     } else {
00461                         r__1 = -poles[i__ + (poles_dim1 << 1)];
00462                         rwork[i__] = z__[j] / (slamc3_(&dsigj, &r__1) - difl[
00463                                 i__]) / (dsigj + poles[i__ + poles_dim1]) / 
00464                                 difr[i__ + (difr_dim1 << 1)];
00465                     }
00466 /* L120: */
00467                 }
00468 
00469 /*              Since B and BX are complex, the following call to SGEMV */
00470 /*              is performed in two steps (real and imaginary parts). */
00471 
00472 /*              CALL SGEMV( 'T', K, NRHS, ONE, B, LDB, WORK, 1, ZERO, */
00473 /*    $                     BX( J, 1 ), LDBX ) */
00474 
00475                 i__ = *k + (*nrhs << 1);
00476                 i__2 = *nrhs;
00477                 for (jcol = 1; jcol <= i__2; ++jcol) {
00478                     i__3 = *k;
00479                     for (jrow = 1; jrow <= i__3; ++jrow) {
00480                         ++i__;
00481                         i__4 = jrow + jcol * b_dim1;
00482                         rwork[i__] = b[i__4].r;
00483 /* L130: */
00484                     }
00485 /* L140: */
00486                 }
00487                 sgemv_("T", k, nrhs, &c_b13, &rwork[*k + 1 + (*nrhs << 1)], k, 
00488                          &rwork[1], &c__1, &c_b15, &rwork[*k + 1], &c__1);
00489                 i__ = *k + (*nrhs << 1);
00490                 i__2 = *nrhs;
00491                 for (jcol = 1; jcol <= i__2; ++jcol) {
00492                     i__3 = *k;
00493                     for (jrow = 1; jrow <= i__3; ++jrow) {
00494                         ++i__;
00495                         rwork[i__] = r_imag(&b[jrow + jcol * b_dim1]);
00496 /* L150: */
00497                     }
00498 /* L160: */
00499                 }
00500                 sgemv_("T", k, nrhs, &c_b13, &rwork[*k + 1 + (*nrhs << 1)], k, 
00501                          &rwork[1], &c__1, &c_b15, &rwork[*k + 1 + *nrhs], &
00502                         c__1);
00503                 i__2 = *nrhs;
00504                 for (jcol = 1; jcol <= i__2; ++jcol) {
00505                     i__3 = j + jcol * bx_dim1;
00506                     i__4 = jcol + *k;
00507                     i__5 = jcol + *k + *nrhs;
00508                     q__1.r = rwork[i__4], q__1.i = rwork[i__5];
00509                     bx[i__3].r = q__1.r, bx[i__3].i = q__1.i;
00510 /* L170: */
00511                 }
00512 /* L180: */
00513             }
00514         }
00515 
00516 /*        Step (2R): if SQRE = 1, apply back the rotation that is */
00517 /*        related to the right null space of the subproblem. */
00518 
00519         if (*sqre == 1) {
00520             ccopy_(nrhs, &b[m + b_dim1], ldb, &bx[m + bx_dim1], ldbx);
00521             csrot_(nrhs, &bx[bx_dim1 + 1], ldbx, &bx[m + bx_dim1], ldbx, c__, 
00522                     s);
00523         }
00524         if (*k < max(m,n)) {
00525             i__1 = n - *k;
00526             clacpy_("A", &i__1, nrhs, &b[*k + 1 + b_dim1], ldb, &bx[*k + 1 + 
00527                     bx_dim1], ldbx);
00528         }
00529 
00530 /*        Step (3R): permute rows of B. */
00531 
00532         ccopy_(nrhs, &bx[bx_dim1 + 1], ldbx, &b[nlp1 + b_dim1], ldb);
00533         if (*sqre == 1) {
00534             ccopy_(nrhs, &bx[m + bx_dim1], ldbx, &b[m + b_dim1], ldb);
00535         }
00536         i__1 = n;
00537         for (i__ = 2; i__ <= i__1; ++i__) {
00538             ccopy_(nrhs, &bx[i__ + bx_dim1], ldbx, &b[perm[i__] + b_dim1], 
00539                     ldb);
00540 /* L190: */
00541         }
00542 
00543 /*        Step (4R): apply back the Givens rotations performed. */
00544 
00545         for (i__ = *givptr; i__ >= 1; --i__) {
00546             r__1 = -givnum[i__ + givnum_dim1];
00547             csrot_(nrhs, &b[givcol[i__ + (givcol_dim1 << 1)] + b_dim1], ldb, &
00548                     b[givcol[i__ + givcol_dim1] + b_dim1], ldb, &givnum[i__ + 
00549                     (givnum_dim1 << 1)], &r__1);
00550 /* L200: */
00551         }
00552     }
00553 
00554     return 0;
00555 
00556 /*     End of CLALS0 */
00557 
00558 } /* clals0_ */


swiftnav
Author(s):
autogenerated on Sat Jun 8 2019 18:55:30