917 lines
28 KiB
C
917 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 int logical;
|
|
typedef short int shortlogical;
|
|
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++ */
|
|
|
|
#define F2C_proc_par_types 1
|
|
#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)
|
|
*/
|
|
|
|
|
|
|
|
|
|
|
|
/* > \brief \b IPARMQ */
|
|
|
|
/* =========== DOCUMENTATION =========== */
|
|
|
|
/* Online html documentation available at */
|
|
/* http://www.netlib.org/lapack/explore-html/ */
|
|
|
|
/* > \htmlonly */
|
|
/* > Download IPARMQ + dependencies */
|
|
/* > <a href="http://www.netlib.org/cgi-bin/netlibfiles.tgz?format=tgz&filename=/lapack/lapack_routine/iparmq.
|
|
f"> */
|
|
/* > [TGZ]</a> */
|
|
/* > <a href="http://www.netlib.org/cgi-bin/netlibfiles.zip?format=zip&filename=/lapack/lapack_routine/iparmq.
|
|
f"> */
|
|
/* > [ZIP]</a> */
|
|
/* > <a href="http://www.netlib.org/cgi-bin/netlibfiles.txt?format=txt&filename=/lapack/lapack_routine/iparmq.
|
|
f"> */
|
|
/* > [TXT]</a> */
|
|
/* > \endhtmlonly */
|
|
|
|
/* Definition: */
|
|
/* =========== */
|
|
|
|
/* INTEGER FUNCTION IPARMQ( ISPEC, NAME, OPTS, N, ILO, IHI, LWORK ) */
|
|
|
|
/* INTEGER IHI, ILO, ISPEC, LWORK, N */
|
|
/* CHARACTER NAME*( * ), OPTS*( * ) */
|
|
|
|
|
|
/* > \par Purpose: */
|
|
/* ============= */
|
|
/* > */
|
|
/* > \verbatim */
|
|
/* > */
|
|
/* > This program sets problem and machine dependent parameters */
|
|
/* > useful for xHSEQR and related subroutines for eigenvalue */
|
|
/* > problems. It is called whenever */
|
|
/* > IPARMQ is called with 12 <= ISPEC <= 16 */
|
|
/* > \endverbatim */
|
|
|
|
/* Arguments: */
|
|
/* ========== */
|
|
|
|
/* > \param[in] ISPEC */
|
|
/* > \verbatim */
|
|
/* > ISPEC is INTEGER */
|
|
/* > ISPEC specifies which tunable parameter IPARMQ should */
|
|
/* > return. */
|
|
/* > */
|
|
/* > ISPEC=12: (INMIN) Matrices of order nmin or less */
|
|
/* > are sent directly to xLAHQR, the implicit */
|
|
/* > double shift QR algorithm. NMIN must be */
|
|
/* > at least 11. */
|
|
/* > */
|
|
/* > ISPEC=13: (INWIN) Size of the deflation window. */
|
|
/* > This is best set greater than or equal to */
|
|
/* > the number of simultaneous shifts NS. */
|
|
/* > Larger matrices benefit from larger deflation */
|
|
/* > windows. */
|
|
/* > */
|
|
/* > ISPEC=14: (INIBL) Determines when to stop nibbling and */
|
|
/* > invest in an (expensive) multi-shift QR sweep. */
|
|
/* > If the aggressive early deflation subroutine */
|
|
/* > finds LD converged eigenvalues from an order */
|
|
/* > NW deflation window and LD > (NW*NIBBLE)/100, */
|
|
/* > then the next QR sweep is skipped and early */
|
|
/* > deflation is applied immediately to the */
|
|
/* > remaining active diagonal block. Setting */
|
|
/* > IPARMQ(ISPEC=14) = 0 causes TTQRE to skip a */
|
|
/* > multi-shift QR sweep whenever early deflation */
|
|
/* > finds a converged eigenvalue. Setting */
|
|
/* > IPARMQ(ISPEC=14) greater than or equal to 100 */
|
|
/* > prevents TTQRE from skipping a multi-shift */
|
|
/* > QR sweep. */
|
|
/* > */
|
|
/* > ISPEC=15: (NSHFTS) The number of simultaneous shifts in */
|
|
/* > a multi-shift QR iteration. */
|
|
/* > */
|
|
/* > ISPEC=16: (IACC22) IPARMQ is set to 0, 1 or 2 with the */
|
|
/* > following meanings. */
|
|
/* > 0: During the multi-shift QR/QZ sweep, */
|
|
/* > blocked eigenvalue reordering, blocked */
|
|
/* > Hessenberg-triangular reduction, */
|
|
/* > reflections and/or rotations are not */
|
|
/* > accumulated when updating the */
|
|
/* > far-from-diagonal matrix entries. */
|
|
/* > 1: During the multi-shift QR/QZ sweep, */
|
|
/* > blocked eigenvalue reordering, blocked */
|
|
/* > Hessenberg-triangular reduction, */
|
|
/* > reflections and/or rotations are */
|
|
/* > accumulated, and matrix-matrix */
|
|
/* > multiplication is used to update the */
|
|
/* > far-from-diagonal matrix entries. */
|
|
/* > 2: During the multi-shift QR/QZ sweep, */
|
|
/* > blocked eigenvalue reordering, blocked */
|
|
/* > Hessenberg-triangular reduction, */
|
|
/* > reflections and/or rotations are */
|
|
/* > accumulated, and 2-by-2 block structure */
|
|
/* > is exploited during matrix-matrix */
|
|
/* > multiplies. */
|
|
/* > (If xTRMM is slower than xGEMM, then */
|
|
/* > IPARMQ(ISPEC=16)=1 may be more efficient than */
|
|
/* > IPARMQ(ISPEC=16)=2 despite the greater level of */
|
|
/* > arithmetic work implied by the latter choice.) */
|
|
/* > \endverbatim */
|
|
/* > */
|
|
/* > \param[in] NAME */
|
|
/* > \verbatim */
|
|
/* > NAME is CHARACTER string */
|
|
/* > Name of the calling subroutine */
|
|
/* > \endverbatim */
|
|
/* > */
|
|
/* > \param[in] OPTS */
|
|
/* > \verbatim */
|
|
/* > OPTS is CHARACTER string */
|
|
/* > This is a concatenation of the string arguments to */
|
|
/* > TTQRE. */
|
|
/* > \endverbatim */
|
|
/* > */
|
|
/* > \param[in] N */
|
|
/* > \verbatim */
|
|
/* > N is INTEGER */
|
|
/* > N is the order of the Hessenberg matrix H. */
|
|
/* > \endverbatim */
|
|
/* > */
|
|
/* > \param[in] ILO */
|
|
/* > \verbatim */
|
|
/* > ILO is INTEGER */
|
|
/* > \endverbatim */
|
|
/* > */
|
|
/* > \param[in] IHI */
|
|
/* > \verbatim */
|
|
/* > IHI is INTEGER */
|
|
/* > It is assumed that H is already upper triangular */
|
|
/* > in rows and columns 1:ILO-1 and IHI+1:N. */
|
|
/* > \endverbatim */
|
|
/* > */
|
|
/* > \param[in] LWORK */
|
|
/* > \verbatim */
|
|
/* > LWORK is INTEGER */
|
|
/* > The amount of workspace available. */
|
|
/* > \endverbatim */
|
|
|
|
/* Authors: */
|
|
/* ======== */
|
|
|
|
/* > \author Univ. of Tennessee */
|
|
/* > \author Univ. of California Berkeley */
|
|
/* > \author Univ. of Colorado Denver */
|
|
/* > \author NAG Ltd. */
|
|
|
|
/* > \date June 2017 */
|
|
|
|
/* > \ingroup OTHERauxiliary */
|
|
|
|
/* > \par Further Details: */
|
|
/* ===================== */
|
|
/* > */
|
|
/* > \verbatim */
|
|
/* > */
|
|
/* > Little is known about how best to choose these parameters. */
|
|
/* > It is possible to use different values of the parameters */
|
|
/* > for each of CHSEQR, DHSEQR, SHSEQR and ZHSEQR. */
|
|
/* > */
|
|
/* > It is probably best to choose different parameters for */
|
|
/* > different matrices and different parameters at different */
|
|
/* > times during the iteration, but this has not been */
|
|
/* > implemented --- yet. */
|
|
/* > */
|
|
/* > */
|
|
/* > The best choices of most of the parameters depend */
|
|
/* > in an ill-understood way on the relative execution */
|
|
/* > rate of xLAQR3 and xLAQR5 and on the nature of each */
|
|
/* > particular eigenvalue problem. Experiment may be the */
|
|
/* > only practical way to determine which choices are most */
|
|
/* > effective. */
|
|
/* > */
|
|
/* > Following is a list of default values supplied by IPARMQ. */
|
|
/* > These defaults may be adjusted in order to attain better */
|
|
/* > performance in any particular computational environment. */
|
|
/* > */
|
|
/* > IPARMQ(ISPEC=12) The xLAHQR vs xLAQR0 crossover point. */
|
|
/* > Default: 75. (Must be at least 11.) */
|
|
/* > */
|
|
/* > IPARMQ(ISPEC=13) Recommended deflation window size. */
|
|
/* > This depends on ILO, IHI and NS, the */
|
|
/* > number of simultaneous shifts returned */
|
|
/* > by IPARMQ(ISPEC=15). The default for */
|
|
/* > (IHI-ILO+1) <= 500 is NS. The default */
|
|
/* > for (IHI-ILO+1) > 500 is 3*NS/2. */
|
|
/* > */
|
|
/* > IPARMQ(ISPEC=14) Nibble crossover point. Default: 14. */
|
|
/* > */
|
|
/* > IPARMQ(ISPEC=15) Number of simultaneous shifts, NS. */
|
|
/* > a multi-shift QR iteration. */
|
|
/* > */
|
|
/* > If IHI-ILO+1 is ... */
|
|
/* > */
|
|
/* > greater than ...but less ... the */
|
|
/* > or equal to ... than default is */
|
|
/* > */
|
|
/* > 0 30 NS = 2+ */
|
|
/* > 30 60 NS = 4+ */
|
|
/* > 60 150 NS = 10 */
|
|
/* > 150 590 NS = ** */
|
|
/* > 590 3000 NS = 64 */
|
|
/* > 3000 6000 NS = 128 */
|
|
/* > 6000 infinity NS = 256 */
|
|
/* > */
|
|
/* > (+) By default matrices of this order are */
|
|
/* > passed to the implicit double shift routine */
|
|
/* > xLAHQR. See IPARMQ(ISPEC=12) above. These */
|
|
/* > values of NS are used only in case of a rare */
|
|
/* > xLAHQR failure. */
|
|
/* > */
|
|
/* > (**) The asterisks (**) indicate an ad-hoc */
|
|
/* > function increasing from 10 to 64. */
|
|
/* > */
|
|
/* > IPARMQ(ISPEC=16) Select structured matrix multiply. */
|
|
/* > (See ISPEC=16 above for details.) */
|
|
/* > Default: 3. */
|
|
/* > \endverbatim */
|
|
/* > */
|
|
/* ===================================================================== */
|
|
integer iparmq_(integer *ispec, char *name__, char *opts, integer *n, integer
|
|
*ilo, integer *ihi, integer *lwork)
|
|
{
|
|
/* System generated locals */
|
|
integer ret_val, i__1, i__2;
|
|
real r__1;
|
|
|
|
/* Local variables */
|
|
integer i__, ic, nh, ns, iz;
|
|
char subnam[6];
|
|
integer name_len;
|
|
|
|
/* -- LAPACK auxiliary routine (version 3.7.1) -- */
|
|
/* -- LAPACK is a software package provided by Univ. of Tennessee, -- */
|
|
/* -- Univ. of California Berkeley, Univ. of Colorado Denver and NAG Ltd..-- */
|
|
/* June 2017 */
|
|
|
|
|
|
/* ================================================================ */
|
|
if (*ispec == 15 || *ispec == 13 || *ispec == 16) {
|
|
|
|
/* ==== Set the number simultaneous shifts ==== */
|
|
|
|
nh = *ihi - *ilo + 1;
|
|
ns = 2;
|
|
if (nh >= 30) {
|
|
ns = 4;
|
|
}
|
|
if (nh >= 60) {
|
|
ns = 10;
|
|
}
|
|
if (nh >= 150) {
|
|
/* Computing MAX */
|
|
r__1 = log((real) nh) / log(2.f);
|
|
i__1 = 10, i__2 = nh / i_nint(&r__1);
|
|
ns = f2cmax(i__1,i__2);
|
|
}
|
|
if (nh >= 590) {
|
|
ns = 64;
|
|
}
|
|
if (nh >= 3000) {
|
|
ns = 128;
|
|
}
|
|
if (nh >= 6000) {
|
|
ns = 256;
|
|
}
|
|
/* Computing MAX */
|
|
i__1 = 2, i__2 = ns - ns % 2;
|
|
ns = f2cmax(i__1,i__2);
|
|
}
|
|
|
|
if (*ispec == 12) {
|
|
|
|
|
|
/* ===== Matrices of order smaller than NMIN get sent */
|
|
/* . to xLAHQR, the classic double shift algorithm. */
|
|
/* . This must be at least 11. ==== */
|
|
|
|
ret_val = 75;
|
|
|
|
} else if (*ispec == 14) {
|
|
|
|
/* ==== INIBL: skip a multi-shift qr iteration and */
|
|
/* . whenever aggressive early deflation finds */
|
|
/* . at least (NIBBLE*(window size)/100) deflations. ==== */
|
|
|
|
ret_val = 14;
|
|
|
|
} else if (*ispec == 15) {
|
|
|
|
/* ==== NSHFTS: The number of simultaneous shifts ===== */
|
|
|
|
ret_val = ns;
|
|
|
|
} else if (*ispec == 13) {
|
|
|
|
/* ==== NW: deflation window size. ==== */
|
|
|
|
if (nh <= 500) {
|
|
ret_val = ns;
|
|
} else {
|
|
ret_val = ns * 3 / 2;
|
|
}
|
|
|
|
} else if (*ispec == 16) {
|
|
|
|
/* ==== IACC22: Whether to accumulate reflections */
|
|
/* . before updating the far-from-diagonal elements */
|
|
/* . and whether to use 2-by-2 block structure while */
|
|
/* . doing it. A small amount of work could be saved */
|
|
/* . by making this choice dependent also upon the */
|
|
/* . NH=IHI-ILO+1. */
|
|
|
|
|
|
/* Convert NAME to upper case if the first character is lower case. */
|
|
|
|
ret_val = 0;
|
|
s_copy(subnam, name__, (ftnlen)6, name_len);
|
|
ic = *(unsigned char *)subnam;
|
|
iz = 'Z';
|
|
if (iz == 90 || iz == 122) {
|
|
|
|
/* ASCII character set */
|
|
|
|
if (ic >= 97 && ic <= 122) {
|
|
*(unsigned char *)subnam = (char) (ic - 32);
|
|
for (i__ = 2; i__ <= 6; ++i__) {
|
|
ic = *(unsigned char *)&subnam[i__ - 1];
|
|
if (ic >= 97 && ic <= 122) {
|
|
*(unsigned char *)&subnam[i__ - 1] = (char) (ic - 32);
|
|
}
|
|
}
|
|
}
|
|
|
|
} else if (iz == 233 || iz == 169) {
|
|
|
|
/* EBCDIC character set */
|
|
|
|
if (ic >= 129 && ic <= 137 || ic >= 145 && ic <= 153 || ic >= 162
|
|
&& ic <= 169) {
|
|
*(unsigned char *)subnam = (char) (ic + 64);
|
|
for (i__ = 2; i__ <= 6; ++i__) {
|
|
ic = *(unsigned char *)&subnam[i__ - 1];
|
|
if (ic >= 129 && ic <= 137 || ic >= 145 && ic <= 153 ||
|
|
ic >= 162 && ic <= 169) {
|
|
*(unsigned char *)&subnam[i__ - 1] = (char) (ic + 64);
|
|
}
|
|
}
|
|
}
|
|
|
|
} else if (iz == 218 || iz == 250) {
|
|
|
|
/* Prime machines: ASCII+128 */
|
|
|
|
if (ic >= 225 && ic <= 250) {
|
|
*(unsigned char *)subnam = (char) (ic - 32);
|
|
for (i__ = 2; i__ <= 6; ++i__) {
|
|
ic = *(unsigned char *)&subnam[i__ - 1];
|
|
if (ic >= 225 && ic <= 250) {
|
|
*(unsigned char *)&subnam[i__ - 1] = (char) (ic - 32);
|
|
}
|
|
}
|
|
}
|
|
}
|
|
|
|
if (s_cmp(subnam + 1, "GGHRD", (ftnlen)5, (ftnlen)5) == 0 || s_cmp(
|
|
subnam + 1, "GGHD3", (ftnlen)5, (ftnlen)5) == 0) {
|
|
ret_val = 1;
|
|
if (nh >= 14) {
|
|
ret_val = 2;
|
|
}
|
|
} else if (s_cmp(subnam + 3, "EXC", (ftnlen)3, (ftnlen)3) == 0) {
|
|
if (nh >= 14) {
|
|
ret_val = 1;
|
|
}
|
|
if (nh >= 14) {
|
|
ret_val = 2;
|
|
}
|
|
} else if (s_cmp(subnam + 1, "HSEQR", (ftnlen)5, (ftnlen)5) == 0 ||
|
|
s_cmp(subnam + 1, "LAQR", (ftnlen)4, (ftnlen)4) == 0) {
|
|
if (ns >= 14) {
|
|
ret_val = 1;
|
|
}
|
|
if (ns >= 14) {
|
|
ret_val = 2;
|
|
}
|
|
}
|
|
|
|
} else {
|
|
/* ===== invalid value of ispec ===== */
|
|
ret_val = -1;
|
|
|
|
}
|
|
|
|
/* ==== End of IPARMQ ==== */
|
|
|
|
return ret_val;
|
|
} /* iparmq_ */
|
|
|