dormqr.c
来自「InsightToolkit-1.4.0(有大量的优化算法程序)」· C语言 代码 · 共 277 行
C
277 行
#include "f2c.h"
#include "netlib.h"
/* Table of constant values */
static integer c__1 = 1;
static integer c_n1 = -1;
static integer c__2 = 2;
static ftnlen cc__2 = 2;
static integer c__65 = 65;
/* Subroutine */ void dormqr_(side, trans, m, n, k, a, lda, tau, c, ldc, work, lwork, info)
const char *side, *trans;
const integer *m, *n, *k;
doublereal *a;
const integer *lda;
doublereal *tau, *c;
const integer *ldc;
doublereal *work;
integer *lwork, *info;
{
/* System generated locals */
address a__1[2];
integer a_dim1, a_offset, c_dim1, c_offset, i__1;
ftnlen i__2[2];
char ch__1[2];
/* Local variables */
static logical left;
static integer i;
static doublereal t[4160] /* was [65][64] */;
static integer nbmin, iinfo, i1, i2, i3;
static integer ib, ic, jc, nb, mi, ni;
static integer nq, nw;
static logical notran;
static integer ldwork, lwkopt;
static logical lquery;
static integer iws;
/* -- LAPACK routine (version 3.0) -- */
/* Univ. of Tennessee, Univ. of California Berkeley, NAG Ltd., */
/* Courant Institute, Argonne National Lab, and Rice University */
/* June 30, 1999 */
/* */
/* Purpose */
/* ======= */
/* */
/* DORMQR overwrites the general real M-by-N matrix C with */
/* */
/* SIDE = 'L' SIDE = 'R' */
/* TRANS = 'N': Q * C C * Q */
/* TRANS = 'T': Q**T * C C * Q**T */
/* */
/* where Q is a real orthogonal matrix defined as the product of k */
/* elementary reflectors */
/* */
/* Q = H(1) H(2) . . . H(k) */
/* */
/* as returned by DGEQRF. Q is of order M if SIDE = 'L' and of order N */
/* if SIDE = 'R'. */
/* */
/* Arguments */
/* ========= */
/* */
/* SIDE (input) CHARACTER*1 */
/* = 'L': apply Q or Q**T from the Left; */
/* = 'R': apply Q or Q**T from the Right. */
/* */
/* TRANS (input) CHARACTER*1 */
/* = 'N': No transpose, apply Q; */
/* = 'T': Transpose, apply Q**T. */
/* */
/* M (input) INTEGER */
/* The number of rows of the matrix C. M >= 0. */
/* */
/* N (input) INTEGER */
/* The number of columns of the matrix C. N >= 0. */
/* */
/* K (input) INTEGER */
/* The number of elementary reflectors whose product defines */
/* the matrix Q. */
/* If SIDE = 'L', M >= K >= 0; */
/* if SIDE = 'R', N >= K >= 0. */
/* */
/* A (input) DOUBLE PRECISION array, dimension (LDA,K) */
/* The i-th column must contain the vector which defines the */
/* elementary reflector H(i), for i = 1,2,...,k, as returned by */
/* DGEQRF in the first k columns of its array argument A. */
/* A is modified by the routine but restored on exit. */
/* */
/* LDA (input) INTEGER */
/* The leading dimension of the array A. */
/* If SIDE = 'L', LDA >= max(1,M); */
/* if SIDE = 'R', LDA >= max(1,N). */
/* */
/* TAU (input) DOUBLE PRECISION array, dimension (K) */
/* TAU(i) must contain the scalar factor of the elementary */
/* reflector H(i), as returned by DGEQRF. */
/* */
/* C (input/output) DOUBLE PRECISION array, dimension (LDC,N) */
/* On entry, the M-by-N matrix C. */
/* On exit, C is overwritten by Q*C or Q**T*C or C*Q**T or C*Q. */
/* */
/* LDC (input) INTEGER */
/* The leading dimension of the array C. LDC >= max(1,M). */
/* */
/* WORK (workspace/output) DOUBLE PRECISION array, dimension (LWORK) */
/* On exit, if INFO = 0, WORK(1) returns the optimal LWORK. */
/* */
/* LWORK (input) INTEGER */
/* The dimension of the array WORK. */
/* If SIDE = 'L', LWORK >= max(1,N); */
/* if SIDE = 'R', LWORK >= max(1,M). */
/* For optimum performance LWORK >= N*NB if SIDE = 'L', and */
/* LWORK >= M*NB if SIDE = 'R', where NB is the optimal */
/* blocksize. */
/* */
/* If LWORK = -1, then a workspace query is assumed; the routine */
/* only calculates the optimal size of the WORK array, returns */
/* this value as the first entry of the WORK array, and no error */
/* message related to LWORK is issued by XERBLA. */
/* */
/* INFO (output) INTEGER */
/* = 0: successful exit */
/* < 0: if INFO = -i, the i-th argument had an illegal value */
/* */
/* ===================================================================== */
/* Parameter adjustments */
a_dim1 = *lda;
a_offset = 1 + a_dim1 * 1;
a -= a_offset;
--tau;
c_dim1 = *ldc;
c_offset = 1 + c_dim1 * 1;
c -= c_offset;
--work;
/* Test the input arguments */
*info = 0;
left = lsame_(side, "L");
notran = lsame_(trans, "N");
lquery = *lwork == -1;
/* NQ is the order of Q and NW is the minimum dimension of WORK */
if (left) {
nq = *m;
nw = *n;
} else {
nq = *n;
nw = *m;
}
if (! left && ! lsame_(side, "R")) {
*info = -1;
} else if (! notran && ! lsame_(trans, "T")) {
*info = -2;
} else if (*m < 0) {
*info = -3;
} else if (*n < 0) {
*info = -4;
} else if (*k < 0 || *k > nq) {
*info = -5;
} else if (*lda < max(1,nq)) {
*info = -7;
} else if (*ldc < max(1,*m)) {
*info = -10;
} else if (*lwork < max(1,nw) && ! lquery) {
*info = -12;
}
if (*info == 0) {
/* Determine the block size. NB may be at most NBMAX, where NBMAX */
/* is used to define the local array T. */
/* Writing concatenation */
i__2[0] = 1, a__1[0] = side;
i__2[1] = 1, a__1[1] = trans;
s_cat(ch__1, a__1, i__2, &cc__2, (ftnlen)2);
nb = ilaenv_(&c__1, "DORMQR", ch__1, m, n, k, &c_n1);
nb = min(64,nb);
lwkopt = max(1,nw) * nb;
work[1] = (doublereal) lwkopt;
}
if (*info != 0) {
i__1 = -(*info);
xerbla_("DORMQR", &i__1);
return;
} else if (lquery) {
return;
}
/* Quick return if possible */
if (*m == 0 || *n == 0 || *k == 0) {
work[1] = 1.;
return;
}
nbmin = 2;
ldwork = nw;
if (nb > 1 && nb < *k) {
iws = nw * nb;
if (*lwork < iws) {
nb = *lwork / ldwork;
/* Writing concatenation */
i__2[0] = 1, a__1[0] = side;
i__2[1] = 1, a__1[1] = trans;
s_cat(ch__1, a__1, i__2, &cc__2, (ftnlen)2);
nbmin = ilaenv_(&c__2, "DORMQR", ch__1, m, n, k, &c_n1);
nbmin = max(2,nbmin);
}
} else {
iws = nw;
}
if (nb < nbmin || nb >= *k) {
/* Use unblocked code */
dorm2r_(side, trans, m, n, k, &a[a_offset], lda, &tau[1], &c[c_offset], ldc, &work[1], &iinfo);
} else {
/* Use blocked code */
if ( (left && ! notran) || (! left && notran) ) {
i1 = 1;
i2 = *k;
i3 = nb;
} else {
i1 = (*k - 1) / nb * nb + 1;
i2 = 1;
i3 = -nb;
}
if (left) {
ni = *n;
jc = 1;
} else {
mi = *m;
ic = 1;
}
for (i = i1; i3 < 0 ? i >= i2 : i <= i2; i += i3) {
ib = min(nb, *k - i + 1);
/* Form the triangular factor of the block reflector */
/* H = H(i) H(i+1) . . . H(i+ib-1) */
i__1 = nq - i + 1;
dlarft_("Forward", "Columnwise", &i__1, &ib, &a[i + i * a_dim1], lda, &tau[i], t, &c__65);
if (left) {
/* H or H' is applied to C(i:m,1:n) */
mi = *m - i + 1;
ic = i;
} else {
/* H or H' is applied to C(1:m,i:n) */
ni = *n - i + 1;
jc = i;
}
/* Apply H or H' */
dlarfb_(side, trans, "Forward", "Columnwise", &mi, &ni, &ib,
&a[i + i * a_dim1], lda, t, &c__65, &c[ic + jc * c_dim1], ldc, &work[1], &ldwork);
}
}
work[1] = (doublereal) lwkopt;
} /* dormqr_ */
⌨️ 快捷键说明
复制代码Ctrl + C
搜索代码Ctrl + F
全屏模式F11
增大字号Ctrl + =
减小字号Ctrl + -
显示快捷键?