958 lines
28 KiB
C
958 lines
28 KiB
C
#include <math.h>
|
|
#include <stdlib.h>
|
|
#include <string.h>
|
|
#include <stdio.h>
|
|
#include <complex.h>
|
|
#ifdef complex
|
|
#undef complex
|
|
#endif
|
|
#ifdef I
|
|
#undef I
|
|
#endif
|
|
|
|
#if defined(_WIN64)
|
|
typedef long long BLASLONG;
|
|
typedef unsigned long long BLASULONG;
|
|
#else
|
|
typedef long BLASLONG;
|
|
typedef unsigned long BLASULONG;
|
|
#endif
|
|
|
|
#ifdef LAPACK_ILP64
|
|
typedef BLASLONG blasint;
|
|
#if defined(_WIN64)
|
|
#define blasabs(x) llabs(x)
|
|
#else
|
|
#define blasabs(x) labs(x)
|
|
#endif
|
|
#else
|
|
typedef int blasint;
|
|
#define blasabs(x) abs(x)
|
|
#endif
|
|
|
|
typedef blasint integer;
|
|
|
|
typedef unsigned int uinteger;
|
|
typedef char *address;
|
|
typedef short int shortint;
|
|
typedef float real;
|
|
typedef double doublereal;
|
|
typedef struct { real r, i; } complex;
|
|
typedef struct { doublereal r, i; } doublecomplex;
|
|
#ifdef _MSC_VER
|
|
static inline _Fcomplex Cf(complex *z) {_Fcomplex zz={z->r , z->i}; return zz;}
|
|
static inline _Dcomplex Cd(doublecomplex *z) {_Dcomplex zz={z->r , z->i};return zz;}
|
|
static inline _Fcomplex * _pCf(complex *z) {return (_Fcomplex*)z;}
|
|
static inline _Dcomplex * _pCd(doublecomplex *z) {return (_Dcomplex*)z;}
|
|
#else
|
|
static inline _Complex float Cf(complex *z) {return z->r + z->i*_Complex_I;}
|
|
static inline _Complex double Cd(doublecomplex *z) {return z->r + z->i*_Complex_I;}
|
|
static inline _Complex float * _pCf(complex *z) {return (_Complex float*)z;}
|
|
static inline _Complex double * _pCd(doublecomplex *z) {return (_Complex double*)z;}
|
|
#endif
|
|
#define pCf(z) (*_pCf(z))
|
|
#define pCd(z) (*_pCd(z))
|
|
typedef blasint logical;
|
|
|
|
typedef char logical1;
|
|
typedef char integer1;
|
|
|
|
#define TRUE_ (1)
|
|
#define FALSE_ (0)
|
|
|
|
/* Extern is for use with -E */
|
|
#ifndef Extern
|
|
#define Extern extern
|
|
#endif
|
|
|
|
/* I/O stuff */
|
|
|
|
typedef int flag;
|
|
typedef int ftnlen;
|
|
typedef int ftnint;
|
|
|
|
/*external read, write*/
|
|
typedef struct
|
|
{ flag cierr;
|
|
ftnint ciunit;
|
|
flag ciend;
|
|
char *cifmt;
|
|
ftnint cirec;
|
|
} cilist;
|
|
|
|
/*internal read, write*/
|
|
typedef struct
|
|
{ flag icierr;
|
|
char *iciunit;
|
|
flag iciend;
|
|
char *icifmt;
|
|
ftnint icirlen;
|
|
ftnint icirnum;
|
|
} icilist;
|
|
|
|
/*open*/
|
|
typedef struct
|
|
{ flag oerr;
|
|
ftnint ounit;
|
|
char *ofnm;
|
|
ftnlen ofnmlen;
|
|
char *osta;
|
|
char *oacc;
|
|
char *ofm;
|
|
ftnint orl;
|
|
char *oblnk;
|
|
} olist;
|
|
|
|
/*close*/
|
|
typedef struct
|
|
{ flag cerr;
|
|
ftnint cunit;
|
|
char *csta;
|
|
} cllist;
|
|
|
|
/*rewind, backspace, endfile*/
|
|
typedef struct
|
|
{ flag aerr;
|
|
ftnint aunit;
|
|
} alist;
|
|
|
|
/* inquire */
|
|
typedef struct
|
|
{ flag inerr;
|
|
ftnint inunit;
|
|
char *infile;
|
|
ftnlen infilen;
|
|
ftnint *inex; /*parameters in standard's order*/
|
|
ftnint *inopen;
|
|
ftnint *innum;
|
|
ftnint *innamed;
|
|
char *inname;
|
|
ftnlen innamlen;
|
|
char *inacc;
|
|
ftnlen inacclen;
|
|
char *inseq;
|
|
ftnlen inseqlen;
|
|
char *indir;
|
|
ftnlen indirlen;
|
|
char *infmt;
|
|
ftnlen infmtlen;
|
|
char *inform;
|
|
ftnint informlen;
|
|
char *inunf;
|
|
ftnlen inunflen;
|
|
ftnint *inrecl;
|
|
ftnint *innrec;
|
|
char *inblank;
|
|
ftnlen inblanklen;
|
|
} inlist;
|
|
|
|
#define VOID void
|
|
|
|
union Multitype { /* for multiple entry points */
|
|
integer1 g;
|
|
shortint h;
|
|
integer i;
|
|
/* longint j; */
|
|
real r;
|
|
doublereal d;
|
|
complex c;
|
|
doublecomplex z;
|
|
};
|
|
|
|
typedef union Multitype Multitype;
|
|
|
|
struct Vardesc { /* for Namelist */
|
|
char *name;
|
|
char *addr;
|
|
ftnlen *dims;
|
|
int type;
|
|
};
|
|
typedef struct Vardesc Vardesc;
|
|
|
|
struct Namelist {
|
|
char *name;
|
|
Vardesc **vars;
|
|
int nvars;
|
|
};
|
|
typedef struct Namelist Namelist;
|
|
|
|
#define abs(x) ((x) >= 0 ? (x) : -(x))
|
|
#define dabs(x) (fabs(x))
|
|
#define f2cmin(a,b) ((a) <= (b) ? (a) : (b))
|
|
#define f2cmax(a,b) ((a) >= (b) ? (a) : (b))
|
|
#define dmin(a,b) (f2cmin(a,b))
|
|
#define dmax(a,b) (f2cmax(a,b))
|
|
#define bit_test(a,b) ((a) >> (b) & 1)
|
|
#define bit_clear(a,b) ((a) & ~((uinteger)1 << (b)))
|
|
#define bit_set(a,b) ((a) | ((uinteger)1 << (b)))
|
|
|
|
#define abort_() { sig_die("Fortran abort routine called", 1); }
|
|
#define c_abs(z) (cabsf(Cf(z)))
|
|
#define c_cos(R,Z) { pCf(R)=ccos(Cf(Z)); }
|
|
#ifdef _MSC_VER
|
|
#define c_div(c, a, b) {Cf(c)._Val[0] = (Cf(a)._Val[0]/Cf(b)._Val[0]); Cf(c)._Val[1]=(Cf(a)._Val[1]/Cf(b)._Val[1]);}
|
|
#define z_div(c, a, b) {Cd(c)._Val[0] = (Cd(a)._Val[0]/Cd(b)._Val[0]); Cd(c)._Val[1]=(Cd(a)._Val[1]/df(b)._Val[1]);}
|
|
#else
|
|
#define c_div(c, a, b) {pCf(c) = Cf(a)/Cf(b);}
|
|
#define z_div(c, a, b) {pCd(c) = Cd(a)/Cd(b);}
|
|
#endif
|
|
#define c_exp(R, Z) {pCf(R) = cexpf(Cf(Z));}
|
|
#define c_log(R, Z) {pCf(R) = clogf(Cf(Z));}
|
|
#define c_sin(R, Z) {pCf(R) = csinf(Cf(Z));}
|
|
//#define c_sqrt(R, Z) {*(R) = csqrtf(Cf(Z));}
|
|
#define c_sqrt(R, Z) {pCf(R) = csqrtf(Cf(Z));}
|
|
#define d_abs(x) (fabs(*(x)))
|
|
#define d_acos(x) (acos(*(x)))
|
|
#define d_asin(x) (asin(*(x)))
|
|
#define d_atan(x) (atan(*(x)))
|
|
#define d_atn2(x, y) (atan2(*(x),*(y)))
|
|
#define d_cnjg(R, Z) { pCd(R) = conj(Cd(Z)); }
|
|
#define r_cnjg(R, Z) { pCf(R) = conjf(Cf(Z)); }
|
|
#define d_cos(x) (cos(*(x)))
|
|
#define d_cosh(x) (cosh(*(x)))
|
|
#define d_dim(__a, __b) ( *(__a) > *(__b) ? *(__a) - *(__b) : 0.0 )
|
|
#define d_exp(x) (exp(*(x)))
|
|
#define d_imag(z) (cimag(Cd(z)))
|
|
#define r_imag(z) (cimagf(Cf(z)))
|
|
#define d_int(__x) (*(__x)>0 ? floor(*(__x)) : -floor(- *(__x)))
|
|
#define r_int(__x) (*(__x)>0 ? floor(*(__x)) : -floor(- *(__x)))
|
|
#define d_lg10(x) ( 0.43429448190325182765 * log(*(x)) )
|
|
#define r_lg10(x) ( 0.43429448190325182765 * log(*(x)) )
|
|
#define d_log(x) (log(*(x)))
|
|
#define d_mod(x, y) (fmod(*(x), *(y)))
|
|
#define u_nint(__x) ((__x)>=0 ? floor((__x) + .5) : -floor(.5 - (__x)))
|
|
#define d_nint(x) u_nint(*(x))
|
|
#define u_sign(__a,__b) ((__b) >= 0 ? ((__a) >= 0 ? (__a) : -(__a)) : -((__a) >= 0 ? (__a) : -(__a)))
|
|
#define d_sign(a,b) u_sign(*(a),*(b))
|
|
#define r_sign(a,b) u_sign(*(a),*(b))
|
|
#define d_sin(x) (sin(*(x)))
|
|
#define d_sinh(x) (sinh(*(x)))
|
|
#define d_sqrt(x) (sqrt(*(x)))
|
|
#define d_tan(x) (tan(*(x)))
|
|
#define d_tanh(x) (tanh(*(x)))
|
|
#define i_abs(x) abs(*(x))
|
|
#define i_dnnt(x) ((integer)u_nint(*(x)))
|
|
#define i_len(s, n) (n)
|
|
#define i_nint(x) ((integer)u_nint(*(x)))
|
|
#define i_sign(a,b) ((integer)u_sign((integer)*(a),(integer)*(b)))
|
|
#define pow_dd(ap, bp) ( pow(*(ap), *(bp)))
|
|
#define pow_si(B,E) spow_ui(*(B),*(E))
|
|
#define pow_ri(B,E) spow_ui(*(B),*(E))
|
|
#define pow_di(B,E) dpow_ui(*(B),*(E))
|
|
#define pow_zi(p, a, b) {pCd(p) = zpow_ui(Cd(a), *(b));}
|
|
#define pow_ci(p, a, b) {pCf(p) = cpow_ui(Cf(a), *(b));}
|
|
#define pow_zz(R,A,B) {pCd(R) = cpow(Cd(A),*(B));}
|
|
#define s_cat(lpp, rpp, rnp, np, llp) { ftnlen i, nc, ll; char *f__rp, *lp; ll = (llp); lp = (lpp); for(i=0; i < (int)*(np); ++i) { nc = ll; if((rnp)[i] < nc) nc = (rnp)[i]; ll -= nc; f__rp = (rpp)[i]; while(--nc >= 0) *lp++ = *(f__rp)++; } while(--ll >= 0) *lp++ = ' '; }
|
|
#define s_cmp(a,b,c,d) ((integer)strncmp((a),(b),f2cmin((c),(d))))
|
|
#define s_copy(A,B,C,D) { int __i,__m; for (__i=0, __m=f2cmin((C),(D)); __i<__m && (B)[__i] != 0; ++__i) (A)[__i] = (B)[__i]; }
|
|
#define sig_die(s, kill) { exit(1); }
|
|
#define s_stop(s, n) {exit(0);}
|
|
static char junk[] = "\n@(#)LIBF77 VERSION 19990503\n";
|
|
#define z_abs(z) (cabs(Cd(z)))
|
|
#define z_exp(R, Z) {pCd(R) = cexp(Cd(Z));}
|
|
#define z_sqrt(R, Z) {pCd(R) = csqrt(Cd(Z));}
|
|
#define myexit_() break;
|
|
#define mycycle() continue;
|
|
#define myceiling(w) {ceil(w)}
|
|
#define myhuge(w) {HUGE_VAL}
|
|
//#define mymaxloc_(w,s,e,n) {if (sizeof(*(w)) == sizeof(double)) dmaxloc_((w),*(s),*(e),n); else dmaxloc_((w),*(s),*(e),n);}
|
|
#define mymaxloc(w,s,e,n) {dmaxloc_(w,*(s),*(e),n)}
|
|
|
|
/* procedure parameter types for -A and -C++ */
|
|
|
|
|
|
#ifdef __cplusplus
|
|
typedef logical (*L_fp)(...);
|
|
#else
|
|
typedef logical (*L_fp)();
|
|
#endif
|
|
|
|
static float spow_ui(float x, integer n) {
|
|
float pow=1.0; unsigned long int u;
|
|
if(n != 0) {
|
|
if(n < 0) n = -n, x = 1/x;
|
|
for(u = n; ; ) {
|
|
if(u & 01) pow *= x;
|
|
if(u >>= 1) x *= x;
|
|
else break;
|
|
}
|
|
}
|
|
return pow;
|
|
}
|
|
static double dpow_ui(double x, integer n) {
|
|
double pow=1.0; unsigned long int u;
|
|
if(n != 0) {
|
|
if(n < 0) n = -n, x = 1/x;
|
|
for(u = n; ; ) {
|
|
if(u & 01) pow *= x;
|
|
if(u >>= 1) x *= x;
|
|
else break;
|
|
}
|
|
}
|
|
return pow;
|
|
}
|
|
#ifdef _MSC_VER
|
|
static _Fcomplex cpow_ui(complex x, integer n) {
|
|
complex pow={1.0,0.0}; unsigned long int u;
|
|
if(n != 0) {
|
|
if(n < 0) n = -n, x.r = 1/x.r, x.i=1/x.i;
|
|
for(u = n; ; ) {
|
|
if(u & 01) pow.r *= x.r, pow.i *= x.i;
|
|
if(u >>= 1) x.r *= x.r, x.i *= x.i;
|
|
else break;
|
|
}
|
|
}
|
|
_Fcomplex p={pow.r, pow.i};
|
|
return p;
|
|
}
|
|
#else
|
|
static _Complex float cpow_ui(_Complex float x, integer n) {
|
|
_Complex float pow=1.0; unsigned long int u;
|
|
if(n != 0) {
|
|
if(n < 0) n = -n, x = 1/x;
|
|
for(u = n; ; ) {
|
|
if(u & 01) pow *= x;
|
|
if(u >>= 1) x *= x;
|
|
else break;
|
|
}
|
|
}
|
|
return pow;
|
|
}
|
|
#endif
|
|
#ifdef _MSC_VER
|
|
static _Dcomplex zpow_ui(_Dcomplex x, integer n) {
|
|
_Dcomplex pow={1.0,0.0}; unsigned long int u;
|
|
if(n != 0) {
|
|
if(n < 0) n = -n, x._Val[0] = 1/x._Val[0], x._Val[1] =1/x._Val[1];
|
|
for(u = n; ; ) {
|
|
if(u & 01) pow._Val[0] *= x._Val[0], pow._Val[1] *= x._Val[1];
|
|
if(u >>= 1) x._Val[0] *= x._Val[0], x._Val[1] *= x._Val[1];
|
|
else break;
|
|
}
|
|
}
|
|
_Dcomplex p = {pow._Val[0], pow._Val[1]};
|
|
return p;
|
|
}
|
|
#else
|
|
static _Complex double zpow_ui(_Complex double x, integer n) {
|
|
_Complex double pow=1.0; unsigned long int u;
|
|
if(n != 0) {
|
|
if(n < 0) n = -n, x = 1/x;
|
|
for(u = n; ; ) {
|
|
if(u & 01) pow *= x;
|
|
if(u >>= 1) x *= x;
|
|
else break;
|
|
}
|
|
}
|
|
return pow;
|
|
}
|
|
#endif
|
|
static integer pow_ii(integer x, integer n) {
|
|
integer pow; unsigned long int u;
|
|
if (n <= 0) {
|
|
if (n == 0 || x == 1) pow = 1;
|
|
else if (x != -1) pow = x == 0 ? 1/x : 0;
|
|
else n = -n;
|
|
}
|
|
if ((n > 0) || !(n == 0 || x == 1 || x != -1)) {
|
|
u = n;
|
|
for(pow = 1; ; ) {
|
|
if(u & 01) pow *= x;
|
|
if(u >>= 1) x *= x;
|
|
else break;
|
|
}
|
|
}
|
|
return pow;
|
|
}
|
|
static integer dmaxloc_(double *w, integer s, integer e, integer *n)
|
|
{
|
|
double m; integer i, mi;
|
|
for(m=w[s-1], mi=s, i=s+1; i<=e; i++)
|
|
if (w[i-1]>m) mi=i ,m=w[i-1];
|
|
return mi-s+1;
|
|
}
|
|
static integer smaxloc_(float *w, integer s, integer e, integer *n)
|
|
{
|
|
float m; integer i, mi;
|
|
for(m=w[s-1], mi=s, i=s+1; i<=e; i++)
|
|
if (w[i-1]>m) mi=i ,m=w[i-1];
|
|
return mi-s+1;
|
|
}
|
|
static inline void cdotc_(complex *z, integer *n_, complex *x, integer *incx_, complex *y, integer *incy_) {
|
|
integer n = *n_, incx = *incx_, incy = *incy_, i;
|
|
#ifdef _MSC_VER
|
|
_Fcomplex zdotc = {0.0, 0.0};
|
|
if (incx == 1 && incy == 1) {
|
|
for (i=0;i<n;i++) { /* zdotc = zdotc + dconjg(x(i))* y(i) */
|
|
zdotc._Val[0] += conjf(Cf(&x[i]))._Val[0] * Cf(&y[i])._Val[0];
|
|
zdotc._Val[1] += conjf(Cf(&x[i]))._Val[1] * Cf(&y[i])._Val[1];
|
|
}
|
|
} else {
|
|
for (i=0;i<n;i++) { /* zdotc = zdotc + dconjg(x(i))* y(i) */
|
|
zdotc._Val[0] += conjf(Cf(&x[i*incx]))._Val[0] * Cf(&y[i*incy])._Val[0];
|
|
zdotc._Val[1] += conjf(Cf(&x[i*incx]))._Val[1] * Cf(&y[i*incy])._Val[1];
|
|
}
|
|
}
|
|
pCf(z) = zdotc;
|
|
}
|
|
#else
|
|
_Complex float zdotc = 0.0;
|
|
if (incx == 1 && incy == 1) {
|
|
for (i=0;i<n;i++) { /* zdotc = zdotc + dconjg(x(i))* y(i) */
|
|
zdotc += conjf(Cf(&x[i])) * Cf(&y[i]);
|
|
}
|
|
} else {
|
|
for (i=0;i<n;i++) { /* zdotc = zdotc + dconjg(x(i))* y(i) */
|
|
zdotc += conjf(Cf(&x[i*incx])) * Cf(&y[i*incy]);
|
|
}
|
|
}
|
|
pCf(z) = zdotc;
|
|
}
|
|
#endif
|
|
static inline void zdotc_(doublecomplex *z, integer *n_, doublecomplex *x, integer *incx_, doublecomplex *y, integer *incy_) {
|
|
integer n = *n_, incx = *incx_, incy = *incy_, i;
|
|
#ifdef _MSC_VER
|
|
_Dcomplex zdotc = {0.0, 0.0};
|
|
if (incx == 1 && incy == 1) {
|
|
for (i=0;i<n;i++) { /* zdotc = zdotc + dconjg(x(i))* y(i) */
|
|
zdotc._Val[0] += conj(Cd(&x[i]))._Val[0] * Cd(&y[i])._Val[0];
|
|
zdotc._Val[1] += conj(Cd(&x[i]))._Val[1] * Cd(&y[i])._Val[1];
|
|
}
|
|
} else {
|
|
for (i=0;i<n;i++) { /* zdotc = zdotc + dconjg(x(i))* y(i) */
|
|
zdotc._Val[0] += conj(Cd(&x[i*incx]))._Val[0] * Cd(&y[i*incy])._Val[0];
|
|
zdotc._Val[1] += conj(Cd(&x[i*incx]))._Val[1] * Cd(&y[i*incy])._Val[1];
|
|
}
|
|
}
|
|
pCd(z) = zdotc;
|
|
}
|
|
#else
|
|
_Complex double zdotc = 0.0;
|
|
if (incx == 1 && incy == 1) {
|
|
for (i=0;i<n;i++) { /* zdotc = zdotc + dconjg(x(i))* y(i) */
|
|
zdotc += conj(Cd(&x[i])) * Cd(&y[i]);
|
|
}
|
|
} else {
|
|
for (i=0;i<n;i++) { /* zdotc = zdotc + dconjg(x(i))* y(i) */
|
|
zdotc += conj(Cd(&x[i*incx])) * Cd(&y[i*incy]);
|
|
}
|
|
}
|
|
pCd(z) = zdotc;
|
|
}
|
|
#endif
|
|
static inline void cdotu_(complex *z, integer *n_, complex *x, integer *incx_, complex *y, integer *incy_) {
|
|
integer n = *n_, incx = *incx_, incy = *incy_, i;
|
|
#ifdef _MSC_VER
|
|
_Fcomplex zdotc = {0.0, 0.0};
|
|
if (incx == 1 && incy == 1) {
|
|
for (i=0;i<n;i++) { /* zdotc = zdotc + dconjg(x(i))* y(i) */
|
|
zdotc._Val[0] += Cf(&x[i])._Val[0] * Cf(&y[i])._Val[0];
|
|
zdotc._Val[1] += Cf(&x[i])._Val[1] * Cf(&y[i])._Val[1];
|
|
}
|
|
} else {
|
|
for (i=0;i<n;i++) { /* zdotc = zdotc + dconjg(x(i))* y(i) */
|
|
zdotc._Val[0] += Cf(&x[i*incx])._Val[0] * Cf(&y[i*incy])._Val[0];
|
|
zdotc._Val[1] += Cf(&x[i*incx])._Val[1] * Cf(&y[i*incy])._Val[1];
|
|
}
|
|
}
|
|
pCf(z) = zdotc;
|
|
}
|
|
#else
|
|
_Complex float zdotc = 0.0;
|
|
if (incx == 1 && incy == 1) {
|
|
for (i=0;i<n;i++) { /* zdotc = zdotc + dconjg(x(i))* y(i) */
|
|
zdotc += Cf(&x[i]) * Cf(&y[i]);
|
|
}
|
|
} else {
|
|
for (i=0;i<n;i++) { /* zdotc = zdotc + dconjg(x(i))* y(i) */
|
|
zdotc += Cf(&x[i*incx]) * Cf(&y[i*incy]);
|
|
}
|
|
}
|
|
pCf(z) = zdotc;
|
|
}
|
|
#endif
|
|
static inline void zdotu_(doublecomplex *z, integer *n_, doublecomplex *x, integer *incx_, doublecomplex *y, integer *incy_) {
|
|
integer n = *n_, incx = *incx_, incy = *incy_, i;
|
|
#ifdef _MSC_VER
|
|
_Dcomplex zdotc = {0.0, 0.0};
|
|
if (incx == 1 && incy == 1) {
|
|
for (i=0;i<n;i++) { /* zdotc = zdotc + dconjg(x(i))* y(i) */
|
|
zdotc._Val[0] += Cd(&x[i])._Val[0] * Cd(&y[i])._Val[0];
|
|
zdotc._Val[1] += Cd(&x[i])._Val[1] * Cd(&y[i])._Val[1];
|
|
}
|
|
} else {
|
|
for (i=0;i<n;i++) { /* zdotc = zdotc + dconjg(x(i))* y(i) */
|
|
zdotc._Val[0] += Cd(&x[i*incx])._Val[0] * Cd(&y[i*incy])._Val[0];
|
|
zdotc._Val[1] += Cd(&x[i*incx])._Val[1] * Cd(&y[i*incy])._Val[1];
|
|
}
|
|
}
|
|
pCd(z) = zdotc;
|
|
}
|
|
#else
|
|
_Complex double zdotc = 0.0;
|
|
if (incx == 1 && incy == 1) {
|
|
for (i=0;i<n;i++) { /* zdotc = zdotc + dconjg(x(i))* y(i) */
|
|
zdotc += Cd(&x[i]) * Cd(&y[i]);
|
|
}
|
|
} else {
|
|
for (i=0;i<n;i++) { /* zdotc = zdotc + dconjg(x(i))* y(i) */
|
|
zdotc += Cd(&x[i*incx]) * Cd(&y[i*incy]);
|
|
}
|
|
}
|
|
pCd(z) = zdotc;
|
|
}
|
|
#endif
|
|
/* -- translated by f2c (version 20000121).
|
|
You must link the resulting object file with the libraries:
|
|
-lf2c -lm (in that order)
|
|
*/
|
|
|
|
|
|
|
|
|
|
/* Table of constant values */
|
|
|
|
static integer c__2 = 2;
|
|
static integer c__1 = 1;
|
|
static doublereal c_b10 = 1.;
|
|
static doublereal c_b11 = 0.;
|
|
static integer c_n1 = -1;
|
|
|
|
/* > \brief \b DLAED7 used by sstedc. Computes the updated eigensystem of a diagonal matrix after modification
|
|
by a rank-one symmetric matrix. Used when the original matrix is dense. */
|
|
|
|
/* =========== DOCUMENTATION =========== */
|
|
|
|
/* Online html documentation available at */
|
|
/* http://www.netlib.org/lapack/explore-html/ */
|
|
|
|
/* > \htmlonly */
|
|
/* > Download DLAED7 + dependencies */
|
|
/* > <a href="http://www.netlib.org/cgi-bin/netlibfiles.tgz?format=tgz&filename=/lapack/lapack_routine/dlaed7.
|
|
f"> */
|
|
/* > [TGZ]</a> */
|
|
/* > <a href="http://www.netlib.org/cgi-bin/netlibfiles.zip?format=zip&filename=/lapack/lapack_routine/dlaed7.
|
|
f"> */
|
|
/* > [ZIP]</a> */
|
|
/* > <a href="http://www.netlib.org/cgi-bin/netlibfiles.txt?format=txt&filename=/lapack/lapack_routine/dlaed7.
|
|
f"> */
|
|
/* > [TXT]</a> */
|
|
/* > \endhtmlonly */
|
|
|
|
/* Definition: */
|
|
/* =========== */
|
|
|
|
/* SUBROUTINE DLAED7( ICOMPQ, N, QSIZ, TLVLS, CURLVL, CURPBM, D, Q, */
|
|
/* LDQ, INDXQ, RHO, CUTPNT, QSTORE, QPTR, PRMPTR, */
|
|
/* PERM, GIVPTR, GIVCOL, GIVNUM, WORK, IWORK, */
|
|
/* INFO ) */
|
|
|
|
/* INTEGER CURLVL, CURPBM, CUTPNT, ICOMPQ, INFO, LDQ, N, */
|
|
/* $ QSIZ, TLVLS */
|
|
/* DOUBLE PRECISION RHO */
|
|
/* INTEGER GIVCOL( 2, * ), GIVPTR( * ), INDXQ( * ), */
|
|
/* $ IWORK( * ), PERM( * ), PRMPTR( * ), QPTR( * ) */
|
|
/* DOUBLE PRECISION D( * ), GIVNUM( 2, * ), Q( LDQ, * ), */
|
|
/* $ QSTORE( * ), WORK( * ) */
|
|
|
|
|
|
/* > \par Purpose: */
|
|
/* ============= */
|
|
/* > */
|
|
/* > \verbatim */
|
|
/* > */
|
|
/* > DLAED7 computes the updated eigensystem of a diagonal */
|
|
/* > matrix after modification by a rank-one symmetric matrix. This */
|
|
/* > routine is used only for the eigenproblem which requires all */
|
|
/* > eigenvalues and optionally eigenvectors of a dense symmetric matrix */
|
|
/* > that has been reduced to tridiagonal form. DLAED1 handles */
|
|
/* > the case in which all eigenvalues and eigenvectors of a symmetric */
|
|
/* > tridiagonal matrix are desired. */
|
|
/* > */
|
|
/* > T = Q(in) ( D(in) + RHO * Z*Z**T ) Q**T(in) = Q(out) * D(out) * Q**T(out) */
|
|
/* > */
|
|
/* > where Z = Q**Tu, u is a vector of length N with ones in the */
|
|
/* > CUTPNT and CUTPNT + 1 th elements and zeros elsewhere. */
|
|
/* > */
|
|
/* > The eigenvectors of the original matrix are stored in Q, and the */
|
|
/* > eigenvalues are in D. The algorithm consists of three stages: */
|
|
/* > */
|
|
/* > The first stage consists of deflating the size of the problem */
|
|
/* > when there are multiple eigenvalues or if there is a zero in */
|
|
/* > the Z vector. For each such occurrence the dimension of the */
|
|
/* > secular equation problem is reduced by one. This stage is */
|
|
/* > performed by the routine DLAED8. */
|
|
/* > */
|
|
/* > The second stage consists of calculating the updated */
|
|
/* > eigenvalues. This is done by finding the roots of the secular */
|
|
/* > equation via the routine DLAED4 (as called by DLAED9). */
|
|
/* > This routine also calculates the eigenvectors of the current */
|
|
/* > problem. */
|
|
/* > */
|
|
/* > The final stage consists of computing the updated eigenvectors */
|
|
/* > directly using the updated eigenvalues. The eigenvectors for */
|
|
/* > the current problem are multiplied with the eigenvectors from */
|
|
/* > the overall problem. */
|
|
/* > \endverbatim */
|
|
|
|
/* Arguments: */
|
|
/* ========== */
|
|
|
|
/* > \param[in] ICOMPQ */
|
|
/* > \verbatim */
|
|
/* > ICOMPQ is INTEGER */
|
|
/* > = 0: Compute eigenvalues only. */
|
|
/* > = 1: Compute eigenvectors of original dense symmetric matrix */
|
|
/* > also. On entry, Q contains the orthogonal matrix used */
|
|
/* > to reduce the original matrix to tridiagonal form. */
|
|
/* > \endverbatim */
|
|
/* > */
|
|
/* > \param[in] N */
|
|
/* > \verbatim */
|
|
/* > N is INTEGER */
|
|
/* > The dimension of the symmetric tridiagonal matrix. N >= 0. */
|
|
/* > \endverbatim */
|
|
/* > */
|
|
/* > \param[in] QSIZ */
|
|
/* > \verbatim */
|
|
/* > QSIZ is INTEGER */
|
|
/* > The dimension of the orthogonal matrix used to reduce */
|
|
/* > the full matrix to tridiagonal form. QSIZ >= N if ICOMPQ = 1. */
|
|
/* > \endverbatim */
|
|
/* > */
|
|
/* > \param[in] TLVLS */
|
|
/* > \verbatim */
|
|
/* > TLVLS is INTEGER */
|
|
/* > The total number of merging levels in the overall divide and */
|
|
/* > conquer tree. */
|
|
/* > \endverbatim */
|
|
/* > */
|
|
/* > \param[in] CURLVL */
|
|
/* > \verbatim */
|
|
/* > CURLVL is INTEGER */
|
|
/* > The current level in the overall merge routine, */
|
|
/* > 0 <= CURLVL <= TLVLS. */
|
|
/* > \endverbatim */
|
|
/* > */
|
|
/* > \param[in] CURPBM */
|
|
/* > \verbatim */
|
|
/* > CURPBM is INTEGER */
|
|
/* > The current problem in the current level in the overall */
|
|
/* > merge routine (counting from upper left to lower right). */
|
|
/* > \endverbatim */
|
|
/* > */
|
|
/* > \param[in,out] D */
|
|
/* > \verbatim */
|
|
/* > D is DOUBLE PRECISION array, dimension (N) */
|
|
/* > On entry, the eigenvalues of the rank-1-perturbed matrix. */
|
|
/* > On exit, the eigenvalues of the repaired matrix. */
|
|
/* > \endverbatim */
|
|
/* > */
|
|
/* > \param[in,out] Q */
|
|
/* > \verbatim */
|
|
/* > Q is DOUBLE PRECISION array, dimension (LDQ, N) */
|
|
/* > On entry, the eigenvectors of the rank-1-perturbed matrix. */
|
|
/* > On exit, the eigenvectors of the repaired tridiagonal matrix. */
|
|
/* > \endverbatim */
|
|
/* > */
|
|
/* > \param[in] LDQ */
|
|
/* > \verbatim */
|
|
/* > LDQ is INTEGER */
|
|
/* > The leading dimension of the array Q. LDQ >= f2cmax(1,N). */
|
|
/* > \endverbatim */
|
|
/* > */
|
|
/* > \param[out] INDXQ */
|
|
/* > \verbatim */
|
|
/* > INDXQ is INTEGER array, dimension (N) */
|
|
/* > The permutation which will reintegrate the subproblem just */
|
|
/* > solved back into sorted order, i.e., D( INDXQ( I = 1, N ) ) */
|
|
/* > will be in ascending order. */
|
|
/* > \endverbatim */
|
|
/* > */
|
|
/* > \param[in] RHO */
|
|
/* > \verbatim */
|
|
/* > RHO is DOUBLE PRECISION */
|
|
/* > The subdiagonal element used to create the rank-1 */
|
|
/* > modification. */
|
|
/* > \endverbatim */
|
|
/* > */
|
|
/* > \param[in] CUTPNT */
|
|
/* > \verbatim */
|
|
/* > CUTPNT is INTEGER */
|
|
/* > Contains the location of the last eigenvalue in the leading */
|
|
/* > sub-matrix. f2cmin(1,N) <= CUTPNT <= N. */
|
|
/* > \endverbatim */
|
|
/* > */
|
|
/* > \param[in,out] QSTORE */
|
|
/* > \verbatim */
|
|
/* > QSTORE is DOUBLE PRECISION array, dimension (N**2+1) */
|
|
/* > Stores eigenvectors of submatrices encountered during */
|
|
/* > divide and conquer, packed together. QPTR points to */
|
|
/* > beginning of the submatrices. */
|
|
/* > \endverbatim */
|
|
/* > */
|
|
/* > \param[in,out] QPTR */
|
|
/* > \verbatim */
|
|
/* > QPTR is INTEGER array, dimension (N+2) */
|
|
/* > List of indices pointing to beginning of submatrices stored */
|
|
/* > in QSTORE. The submatrices are numbered starting at the */
|
|
/* > bottom left of the divide and conquer tree, from left to */
|
|
/* > right and bottom to top. */
|
|
/* > \endverbatim */
|
|
/* > */
|
|
/* > \param[in] PRMPTR */
|
|
/* > \verbatim */
|
|
/* > PRMPTR is INTEGER array, dimension (N lg N) */
|
|
/* > Contains a list of pointers which indicate where in PERM a */
|
|
/* > level's permutation is stored. PRMPTR(i+1) - PRMPTR(i) */
|
|
/* > indicates the size of the permutation and also the size of */
|
|
/* > the full, non-deflated problem. */
|
|
/* > \endverbatim */
|
|
/* > */
|
|
/* > \param[in] PERM */
|
|
/* > \verbatim */
|
|
/* > PERM is INTEGER array, dimension (N lg N) */
|
|
/* > Contains the permutations (from deflation and sorting) to be */
|
|
/* > applied to each eigenblock. */
|
|
/* > \endverbatim */
|
|
/* > */
|
|
/* > \param[in] GIVPTR */
|
|
/* > \verbatim */
|
|
/* > GIVPTR is INTEGER array, dimension (N lg N) */
|
|
/* > Contains a list of pointers which indicate where in GIVCOL a */
|
|
/* > level's Givens rotations are stored. GIVPTR(i+1) - GIVPTR(i) */
|
|
/* > indicates the number of Givens rotations. */
|
|
/* > \endverbatim */
|
|
/* > */
|
|
/* > \param[in] GIVCOL */
|
|
/* > \verbatim */
|
|
/* > GIVCOL is INTEGER array, dimension (2, N lg N) */
|
|
/* > Each pair of numbers indicates a pair of columns to take place */
|
|
/* > in a Givens rotation. */
|
|
/* > \endverbatim */
|
|
/* > */
|
|
/* > \param[in] GIVNUM */
|
|
/* > \verbatim */
|
|
/* > GIVNUM is DOUBLE PRECISION array, dimension (2, N lg N) */
|
|
/* > Each number indicates the S value to be used in the */
|
|
/* > corresponding Givens rotation. */
|
|
/* > \endverbatim */
|
|
/* > */
|
|
/* > \param[out] WORK */
|
|
/* > \verbatim */
|
|
/* > WORK is DOUBLE PRECISION array, dimension (3*N+2*QSIZ*N) */
|
|
/* > \endverbatim */
|
|
/* > */
|
|
/* > \param[out] IWORK */
|
|
/* > \verbatim */
|
|
/* > IWORK is INTEGER array, dimension (4*N) */
|
|
/* > \endverbatim */
|
|
/* > */
|
|
/* > \param[out] INFO */
|
|
/* > \verbatim */
|
|
/* > INFO is INTEGER */
|
|
/* > = 0: successful exit. */
|
|
/* > < 0: if INFO = -i, the i-th argument had an illegal value. */
|
|
/* > > 0: if INFO = 1, an eigenvalue did not converge */
|
|
/* > \endverbatim */
|
|
|
|
/* Authors: */
|
|
/* ======== */
|
|
|
|
/* > \author Univ. of Tennessee */
|
|
/* > \author Univ. of California Berkeley */
|
|
/* > \author Univ. of Colorado Denver */
|
|
/* > \author NAG Ltd. */
|
|
|
|
/* > \date June 2016 */
|
|
|
|
/* > \ingroup auxOTHERcomputational */
|
|
|
|
/* > \par Contributors: */
|
|
/* ================== */
|
|
/* > */
|
|
/* > Jeff Rutter, Computer Science Division, University of California */
|
|
/* > at Berkeley, USA */
|
|
|
|
/* ===================================================================== */
|
|
/* Subroutine */ void dlaed7_(integer *icompq, integer *n, integer *qsiz,
|
|
integer *tlvls, integer *curlvl, integer *curpbm, doublereal *d__,
|
|
doublereal *q, integer *ldq, integer *indxq, doublereal *rho, integer
|
|
*cutpnt, doublereal *qstore, integer *qptr, integer *prmptr, integer *
|
|
perm, integer *givptr, integer *givcol, doublereal *givnum,
|
|
doublereal *work, integer *iwork, integer *info)
|
|
{
|
|
/* System generated locals */
|
|
integer q_dim1, q_offset, i__1, i__2;
|
|
|
|
/* Local variables */
|
|
integer indx, curr, i__, k;
|
|
extern /* Subroutine */ void dgemm_(char *, char *, integer *, integer *,
|
|
integer *, doublereal *, doublereal *, integer *, doublereal *,
|
|
integer *, doublereal *, doublereal *, integer *);
|
|
integer indxc, indxp, n1, n2;
|
|
extern /* Subroutine */ void dlaed8_(integer *, integer *, integer *,
|
|
integer *, doublereal *, doublereal *, integer *, integer *,
|
|
doublereal *, integer *, doublereal *, doublereal *, doublereal *,
|
|
integer *, doublereal *, integer *, integer *, integer *,
|
|
doublereal *, integer *, integer *, integer *), dlaed9_(integer *,
|
|
integer *, integer *, integer *, doublereal *, doublereal *,
|
|
integer *, doublereal *, doublereal *, doublereal *, doublereal *,
|
|
integer *, integer *), dlaeda_(integer *, integer *, integer *,
|
|
integer *, integer *, integer *, integer *, integer *, doublereal
|
|
*, doublereal *, integer *, doublereal *, doublereal *, integer *)
|
|
;
|
|
integer idlmda, is, iw, iz;
|
|
extern /* Subroutine */ void dlamrg_(integer *, integer *, doublereal *,
|
|
integer *, integer *, integer *);
|
|
extern int xerbla_(char *, integer *, ftnlen);
|
|
integer coltyp, iq2, ptr, ldq2;
|
|
|
|
|
|
/* -- LAPACK computational routine (version 3.7.0) -- */
|
|
/* -- LAPACK is a software package provided by Univ. of Tennessee, -- */
|
|
/* -- Univ. of California Berkeley, Univ. of Colorado Denver and NAG Ltd..-- */
|
|
/* June 2016 */
|
|
|
|
|
|
/* ===================================================================== */
|
|
|
|
|
|
/* Test the input parameters. */
|
|
|
|
/* Parameter adjustments */
|
|
--d__;
|
|
q_dim1 = *ldq;
|
|
q_offset = 1 + q_dim1 * 1;
|
|
q -= q_offset;
|
|
--indxq;
|
|
--qstore;
|
|
--qptr;
|
|
--prmptr;
|
|
--perm;
|
|
--givptr;
|
|
givcol -= 3;
|
|
givnum -= 3;
|
|
--work;
|
|
--iwork;
|
|
|
|
/* Function Body */
|
|
*info = 0;
|
|
|
|
if (*icompq < 0 || *icompq > 1) {
|
|
*info = -1;
|
|
} else if (*n < 0) {
|
|
*info = -2;
|
|
} else if (*icompq == 1 && *qsiz < *n) {
|
|
*info = -3;
|
|
} else if (*ldq < f2cmax(1,*n)) {
|
|
*info = -9;
|
|
} else if (f2cmin(1,*n) > *cutpnt || *n < *cutpnt) {
|
|
*info = -12;
|
|
}
|
|
if (*info != 0) {
|
|
i__1 = -(*info);
|
|
xerbla_("DLAED7", &i__1, (ftnlen)6);
|
|
return;
|
|
}
|
|
|
|
/* Quick return if possible */
|
|
|
|
if (*n == 0) {
|
|
return;
|
|
}
|
|
|
|
/* The following values are for bookkeeping purposes only. They are */
|
|
/* integer pointers which indicate the portion of the workspace */
|
|
/* used by a particular array in DLAED8 and DLAED9. */
|
|
|
|
if (*icompq == 1) {
|
|
ldq2 = *qsiz;
|
|
} else {
|
|
ldq2 = *n;
|
|
}
|
|
|
|
iz = 1;
|
|
idlmda = iz + *n;
|
|
iw = idlmda + *n;
|
|
iq2 = iw + *n;
|
|
is = iq2 + *n * ldq2;
|
|
|
|
indx = 1;
|
|
indxc = indx + *n;
|
|
coltyp = indxc + *n;
|
|
indxp = coltyp + *n;
|
|
|
|
/* Form the z-vector which consists of the last row of Q_1 and the */
|
|
/* first row of Q_2. */
|
|
|
|
ptr = pow_ii(c__2, *tlvls) + 1;
|
|
i__1 = *curlvl - 1;
|
|
for (i__ = 1; i__ <= i__1; ++i__) {
|
|
i__2 = *tlvls - i__;
|
|
ptr += pow_ii(c__2, i__2);
|
|
/* L10: */
|
|
}
|
|
curr = ptr + *curpbm;
|
|
dlaeda_(n, tlvls, curlvl, curpbm, &prmptr[1], &perm[1], &givptr[1], &
|
|
givcol[3], &givnum[3], &qstore[1], &qptr[1], &work[iz], &work[iz
|
|
+ *n], info);
|
|
|
|
/* When solving the final problem, we no longer need the stored data, */
|
|
/* so we will overwrite the data from this level onto the previously */
|
|
/* used storage space. */
|
|
|
|
if (*curlvl == *tlvls) {
|
|
qptr[curr] = 1;
|
|
prmptr[curr] = 1;
|
|
givptr[curr] = 1;
|
|
}
|
|
|
|
/* Sort and Deflate eigenvalues. */
|
|
|
|
dlaed8_(icompq, &k, n, qsiz, &d__[1], &q[q_offset], ldq, &indxq[1], rho,
|
|
cutpnt, &work[iz], &work[idlmda], &work[iq2], &ldq2, &work[iw], &
|
|
perm[prmptr[curr]], &givptr[curr + 1], &givcol[(givptr[curr] << 1)
|
|
+ 1], &givnum[(givptr[curr] << 1) + 1], &iwork[indxp], &iwork[
|
|
indx], info);
|
|
prmptr[curr + 1] = prmptr[curr] + *n;
|
|
givptr[curr + 1] += givptr[curr];
|
|
|
|
/* Solve Secular Equation. */
|
|
|
|
if (k != 0) {
|
|
dlaed9_(&k, &c__1, &k, n, &d__[1], &work[is], &k, rho, &work[idlmda],
|
|
&work[iw], &qstore[qptr[curr]], &k, info);
|
|
if (*info != 0) {
|
|
goto L30;
|
|
}
|
|
if (*icompq == 1) {
|
|
dgemm_("N", "N", qsiz, &k, &k, &c_b10, &work[iq2], &ldq2, &qstore[
|
|
qptr[curr]], &k, &c_b11, &q[q_offset], ldq);
|
|
}
|
|
/* Computing 2nd power */
|
|
i__1 = k;
|
|
qptr[curr + 1] = qptr[curr] + i__1 * i__1;
|
|
|
|
/* Prepare the INDXQ sorting permutation. */
|
|
|
|
n1 = k;
|
|
n2 = *n - k;
|
|
dlamrg_(&n1, &n2, &d__[1], &c__1, &c_n1, &indxq[1]);
|
|
} else {
|
|
qptr[curr + 1] = qptr[curr];
|
|
i__1 = *n;
|
|
for (i__ = 1; i__ <= i__1; ++i__) {
|
|
indxq[i__] = i__;
|
|
/* L20: */
|
|
}
|
|
}
|
|
|
|
L30:
|
|
return;
|
|
|
|
/* End of DLAED7 */
|
|
|
|
} /* dlaed7_ */
|
|
|