123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960616263646566676869707172737475767778798081828384858687888990919293949596979899100101102103104105106107108109110111112113114115116117118119120121122123124125126127128129130131132133134135136137138139140141142143144145146147148149150151152153154155156157158159160161162163164165166167168169170171172173174175176177178179180181182183184185186187188189190191192193194195196197198199200201202203204205206207208209210211212213214215216217218219220221222223224225226227228229230231232233234235236237238239240241242243244245246247248249250251252253254255256257258259260261262263264265 |
- /* dgetri.f -- translated by f2c (version 20061008).
- You must link the resulting object file with libf2c:
- on Microsoft Windows system, link with libf2c.lib;
- on Linux or Unix systems, link with .../path/to/libf2c.a -lm
- or, if you install libf2c.a in a standard place, with -lf2c -lm
- -- in that order, at the end of the command line, as in
- cc *.o -lf2c -lm
- Source for libf2c is in /netlib/f2c/libf2c.zip, e.g.,
- http://www.netlib.org/f2c/libf2c.zip
- */
- #include "f2c.h"
- #include "blaswrap.h"
- /* Table of constant values */
- static integer c__1 = 1;
- static integer c_n1 = -1;
- static integer c__2 = 2;
- static doublereal c_b20 = -1.;
- static doublereal c_b22 = 1.;
- /* Subroutine */ int _starpu_dgetri_(integer *n, doublereal *a, integer *lda, integer
- *ipiv, doublereal *work, integer *lwork, integer *info)
- {
- /* System generated locals */
- integer a_dim1, a_offset, i__1, i__2, i__3;
- /* Local variables */
- integer i__, j, jb, nb, jj, jp, nn, iws;
- extern /* Subroutine */ int _starpu_dgemm_(char *, char *, integer *, integer *,
- integer *, doublereal *, doublereal *, integer *, doublereal *,
- integer *, doublereal *, doublereal *, integer *),
- _starpu_dgemv_(char *, integer *, integer *, doublereal *, doublereal *,
- integer *, doublereal *, integer *, doublereal *, doublereal *,
- integer *);
- integer nbmin;
- extern /* Subroutine */ int _starpu_dswap_(integer *, doublereal *, integer *,
- doublereal *, integer *), _starpu_dtrsm_(char *, char *, char *, char *,
- integer *, integer *, doublereal *, doublereal *, integer *,
- doublereal *, integer *), _starpu_xerbla_(
- char *, integer *);
- extern integer _starpu_ilaenv_(integer *, char *, char *, integer *, integer *,
- integer *, integer *);
- integer ldwork;
- extern /* Subroutine */ int _starpu_dtrtri_(char *, char *, integer *, doublereal
- *, integer *, integer *);
- integer lwkopt;
- logical lquery;
- /* -- LAPACK routine (version 3.2) -- */
- /* Univ. of Tennessee, Univ. of California Berkeley and NAG Ltd.. */
- /* November 2006 */
- /* .. Scalar Arguments .. */
- /* .. */
- /* .. Array Arguments .. */
- /* .. */
- /* Purpose */
- /* ======= */
- /* DGETRI computes the inverse of a matrix using the LU factorization */
- /* computed by DGETRF. */
- /* This method inverts U and then computes inv(A) by solving the system */
- /* inv(A)*L = inv(U) for inv(A). */
- /* Arguments */
- /* ========= */
- /* N (input) INTEGER */
- /* The order of the matrix A. N >= 0. */
- /* A (input/output) DOUBLE PRECISION array, dimension (LDA,N) */
- /* On entry, the factors L and U from the factorization */
- /* A = P*L*U as computed by DGETRF. */
- /* On exit, if INFO = 0, the inverse of the original matrix A. */
- /* LDA (input) INTEGER */
- /* The leading dimension of the array A. LDA >= max(1,N). */
- /* IPIV (input) INTEGER array, dimension (N) */
- /* The pivot indices from DGETRF; for 1<=i<=N, row i of the */
- /* matrix was interchanged with row IPIV(i). */
- /* WORK (workspace/output) DOUBLE PRECISION array, dimension (MAX(1,LWORK)) */
- /* On exit, if INFO=0, then WORK(1) returns the optimal LWORK. */
- /* LWORK (input) INTEGER */
- /* The dimension of the array WORK. LWORK >= max(1,N). */
- /* For optimal performance LWORK >= N*NB, where NB is */
- /* the optimal blocksize returned by ILAENV. */
- /* 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 */
- /* > 0: if INFO = i, U(i,i) is exactly zero; the matrix is */
- /* singular and its inverse could not be computed. */
- /* ===================================================================== */
- /* .. Parameters .. */
- /* .. */
- /* .. Local Scalars .. */
- /* .. */
- /* .. External Functions .. */
- /* .. */
- /* .. External Subroutines .. */
- /* .. */
- /* .. Intrinsic Functions .. */
- /* .. */
- /* .. Executable Statements .. */
- /* Test the input parameters. */
- /* Parameter adjustments */
- a_dim1 = *lda;
- a_offset = 1 + a_dim1;
- a -= a_offset;
- --ipiv;
- --work;
- /* Function Body */
- *info = 0;
- nb = _starpu_ilaenv_(&c__1, "DGETRI", " ", n, &c_n1, &c_n1, &c_n1);
- lwkopt = *n * nb;
- work[1] = (doublereal) lwkopt;
- lquery = *lwork == -1;
- if (*n < 0) {
- *info = -1;
- } else if (*lda < max(1,*n)) {
- *info = -3;
- } else if (*lwork < max(1,*n) && ! lquery) {
- *info = -6;
- }
- if (*info != 0) {
- i__1 = -(*info);
- _starpu_xerbla_("DGETRI", &i__1);
- return 0;
- } else if (lquery) {
- return 0;
- }
- /* Quick return if possible */
- if (*n == 0) {
- return 0;
- }
- /* Form inv(U). If INFO > 0 from DTRTRI, then U is singular, */
- /* and the inverse is not computed. */
- _starpu_dtrtri_("Upper", "Non-unit", n, &a[a_offset], lda, info);
- if (*info > 0) {
- return 0;
- }
- nbmin = 2;
- ldwork = *n;
- if (nb > 1 && nb < *n) {
- /* Computing MAX */
- i__1 = ldwork * nb;
- iws = max(i__1,1);
- if (*lwork < iws) {
- nb = *lwork / ldwork;
- /* Computing MAX */
- i__1 = 2, i__2 = _starpu_ilaenv_(&c__2, "DGETRI", " ", n, &c_n1, &c_n1, &
- c_n1);
- nbmin = max(i__1,i__2);
- }
- } else {
- iws = *n;
- }
- /* Solve the equation inv(A)*L = inv(U) for inv(A). */
- if (nb < nbmin || nb >= *n) {
- /* Use unblocked code. */
- for (j = *n; j >= 1; --j) {
- /* Copy current column of L to WORK and replace with zeros. */
- i__1 = *n;
- for (i__ = j + 1; i__ <= i__1; ++i__) {
- work[i__] = a[i__ + j * a_dim1];
- a[i__ + j * a_dim1] = 0.;
- /* L10: */
- }
- /* Compute current column of inv(A). */
- if (j < *n) {
- i__1 = *n - j;
- _starpu_dgemv_("No transpose", n, &i__1, &c_b20, &a[(j + 1) * a_dim1
- + 1], lda, &work[j + 1], &c__1, &c_b22, &a[j * a_dim1
- + 1], &c__1);
- }
- /* L20: */
- }
- } else {
- /* Use blocked code. */
- nn = (*n - 1) / nb * nb + 1;
- i__1 = -nb;
- for (j = nn; i__1 < 0 ? j >= 1 : j <= 1; j += i__1) {
- /* Computing MIN */
- i__2 = nb, i__3 = *n - j + 1;
- jb = min(i__2,i__3);
- /* Copy current block column of L to WORK and replace with */
- /* zeros. */
- i__2 = j + jb - 1;
- for (jj = j; jj <= i__2; ++jj) {
- i__3 = *n;
- for (i__ = jj + 1; i__ <= i__3; ++i__) {
- work[i__ + (jj - j) * ldwork] = a[i__ + jj * a_dim1];
- a[i__ + jj * a_dim1] = 0.;
- /* L30: */
- }
- /* L40: */
- }
- /* Compute current block column of inv(A). */
- if (j + jb <= *n) {
- i__2 = *n - j - jb + 1;
- _starpu_dgemm_("No transpose", "No transpose", n, &jb, &i__2, &c_b20,
- &a[(j + jb) * a_dim1 + 1], lda, &work[j + jb], &
- ldwork, &c_b22, &a[j * a_dim1 + 1], lda);
- }
- _starpu_dtrsm_("Right", "Lower", "No transpose", "Unit", n, &jb, &c_b22, &
- work[j], &ldwork, &a[j * a_dim1 + 1], lda);
- /* L50: */
- }
- }
- /* Apply column interchanges. */
- for (j = *n - 1; j >= 1; --j) {
- jp = ipiv[j];
- if (jp != j) {
- _starpu_dswap_(n, &a[j * a_dim1 + 1], &c__1, &a[jp * a_dim1 + 1], &c__1);
- }
- /* L60: */
- }
- work[1] = (doublereal) iws;
- return 0;
- /* End of DGETRI */
- } /* _starpu_dgetri_ */
|