904 lines
24 KiB
C
904 lines
24 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)
|
|
*/
|
|
|
|
|
|
|
|
|
|
/* > \brief \b SLA_GEAMV computes a matrix-vector product using a general matrix to calculate error bounds. */
|
|
|
|
/* =========== DOCUMENTATION =========== */
|
|
|
|
/* Online html documentation available at */
|
|
/* http://www.netlib.org/lapack/explore-html/ */
|
|
|
|
/* > \htmlonly */
|
|
/* > Download SLA_GEAMV + dependencies */
|
|
/* > <a href="http://www.netlib.org/cgi-bin/netlibfiles.tgz?format=tgz&filename=/lapack/lapack_routine/sla_gea
|
|
mv.f"> */
|
|
/* > [TGZ]</a> */
|
|
/* > <a href="http://www.netlib.org/cgi-bin/netlibfiles.zip?format=zip&filename=/lapack/lapack_routine/sla_gea
|
|
mv.f"> */
|
|
/* > [ZIP]</a> */
|
|
/* > <a href="http://www.netlib.org/cgi-bin/netlibfiles.txt?format=txt&filename=/lapack/lapack_routine/sla_gea
|
|
mv.f"> */
|
|
/* > [TXT]</a> */
|
|
/* > \endhtmlonly */
|
|
|
|
/* Definition: */
|
|
/* =========== */
|
|
|
|
/* SUBROUTINE SLA_GEAMV ( TRANS, M, N, ALPHA, A, LDA, X, INCX, BETA, */
|
|
/* Y, INCY ) */
|
|
|
|
/* REAL ALPHA, BETA */
|
|
/* INTEGER INCX, INCY, LDA, M, N, TRANS */
|
|
/* REAL A( LDA, * ), X( * ), Y( * ) */
|
|
|
|
|
|
/* > \par Purpose: */
|
|
/* ============= */
|
|
/* > */
|
|
/* > \verbatim */
|
|
/* > */
|
|
/* > SLA_GEAMV performs one of the matrix-vector operations */
|
|
/* > */
|
|
/* > y := alpha*abs(A)*abs(x) + beta*abs(y), */
|
|
/* > or y := alpha*abs(A)**T*abs(x) + beta*abs(y), */
|
|
/* > */
|
|
/* > where alpha and beta are scalars, x and y are vectors and A is an */
|
|
/* > m by n matrix. */
|
|
/* > */
|
|
/* > This function is primarily used in calculating error bounds. */
|
|
/* > To protect against underflow during evaluation, components in */
|
|
/* > the resulting vector are perturbed away from zero by (N+1) */
|
|
/* > times the underflow threshold. To prevent unnecessarily large */
|
|
/* > errors for block-structure embedded in general matrices, */
|
|
/* > "symbolically" zero components are not perturbed. A zero */
|
|
/* > entry is considered "symbolic" if all multiplications involved */
|
|
/* > in computing that entry have at least one zero multiplicand. */
|
|
/* > \endverbatim */
|
|
|
|
/* Arguments: */
|
|
/* ========== */
|
|
|
|
/* > \param[in] TRANS */
|
|
/* > \verbatim */
|
|
/* > TRANS is INTEGER */
|
|
/* > On entry, TRANS specifies the operation to be performed as */
|
|
/* > follows: */
|
|
/* > */
|
|
/* > BLAS_NO_TRANS y := alpha*abs(A)*abs(x) + beta*abs(y) */
|
|
/* > BLAS_TRANS y := alpha*abs(A**T)*abs(x) + beta*abs(y) */
|
|
/* > BLAS_CONJ_TRANS y := alpha*abs(A**T)*abs(x) + beta*abs(y) */
|
|
/* > */
|
|
/* > Unchanged on exit. */
|
|
/* > \endverbatim */
|
|
/* > */
|
|
/* > \param[in] M */
|
|
/* > \verbatim */
|
|
/* > M is INTEGER */
|
|
/* > On entry, M specifies the number of rows of the matrix A. */
|
|
/* > M must be at least zero. */
|
|
/* > Unchanged on exit. */
|
|
/* > \endverbatim */
|
|
/* > */
|
|
/* > \param[in] N */
|
|
/* > \verbatim */
|
|
/* > N is INTEGER */
|
|
/* > On entry, N specifies the number of columns of the matrix A. */
|
|
/* > N must be at least zero. */
|
|
/* > Unchanged on exit. */
|
|
/* > \endverbatim */
|
|
/* > */
|
|
/* > \param[in] ALPHA */
|
|
/* > \verbatim */
|
|
/* > ALPHA is REAL */
|
|
/* > On entry, ALPHA specifies the scalar alpha. */
|
|
/* > Unchanged on exit. */
|
|
/* > \endverbatim */
|
|
/* > */
|
|
/* > \param[in] A */
|
|
/* > \verbatim */
|
|
/* > A is REAL array, dimension ( LDA, n ) */
|
|
/* > Before entry, the leading m by n part of the array A must */
|
|
/* > contain the matrix of coefficients. */
|
|
/* > Unchanged on exit. */
|
|
/* > \endverbatim */
|
|
/* > */
|
|
/* > \param[in] LDA */
|
|
/* > \verbatim */
|
|
/* > LDA is INTEGER */
|
|
/* > On entry, LDA specifies the first dimension of A as declared */
|
|
/* > in the calling (sub) program. LDA must be at least */
|
|
/* > f2cmax( 1, m ). */
|
|
/* > Unchanged on exit. */
|
|
/* > \endverbatim */
|
|
/* > */
|
|
/* > \param[in] X */
|
|
/* > \verbatim */
|
|
/* > X is REAL array, dimension */
|
|
/* > ( 1 + ( n - 1 )*abs( INCX ) ) when TRANS = 'N' or 'n' */
|
|
/* > and at least */
|
|
/* > ( 1 + ( m - 1 )*abs( INCX ) ) otherwise. */
|
|
/* > Before entry, the incremented array X must contain the */
|
|
/* > vector x. */
|
|
/* > Unchanged on exit. */
|
|
/* > \endverbatim */
|
|
/* > */
|
|
/* > \param[in] INCX */
|
|
/* > \verbatim */
|
|
/* > INCX is INTEGER */
|
|
/* > On entry, INCX specifies the increment for the elements of */
|
|
/* > X. INCX must not be zero. */
|
|
/* > Unchanged on exit. */
|
|
/* > \endverbatim */
|
|
/* > */
|
|
/* > \param[in] BETA */
|
|
/* > \verbatim */
|
|
/* > BETA is REAL */
|
|
/* > On entry, BETA specifies the scalar beta. When BETA is */
|
|
/* > supplied as zero then Y need not be set on input. */
|
|
/* > Unchanged on exit. */
|
|
/* > \endverbatim */
|
|
/* > */
|
|
/* > \param[in,out] Y */
|
|
/* > \verbatim */
|
|
/* > Y is REAL array, */
|
|
/* > dimension at least */
|
|
/* > ( 1 + ( m - 1 )*abs( INCY ) ) when TRANS = 'N' or 'n' */
|
|
/* > and at least */
|
|
/* > ( 1 + ( n - 1 )*abs( INCY ) ) otherwise. */
|
|
/* > Before entry with BETA non-zero, the incremented array Y */
|
|
/* > must contain the vector y. On exit, Y is overwritten by the */
|
|
/* > updated vector y. */
|
|
/* > \endverbatim */
|
|
/* > */
|
|
/* > \param[in] INCY */
|
|
/* > \verbatim */
|
|
/* > INCY is INTEGER */
|
|
/* > On entry, INCY specifies the increment for the elements of */
|
|
/* > Y. INCY must not be zero. */
|
|
/* > Unchanged on exit. */
|
|
/* > */
|
|
/* > Level 2 Blas routine. */
|
|
/* > \endverbatim */
|
|
|
|
/* Authors: */
|
|
/* ======== */
|
|
|
|
/* > \author Univ. of Tennessee */
|
|
/* > \author Univ. of California Berkeley */
|
|
/* > \author Univ. of Colorado Denver */
|
|
/* > \author NAG Ltd. */
|
|
|
|
/* > \date June 2017 */
|
|
|
|
/* > \ingroup realGEcomputational */
|
|
|
|
/* ===================================================================== */
|
|
/* Subroutine */ void sla_geamv_(integer *trans, integer *m, integer *n, real
|
|
*alpha, real *a, integer *lda, real *x, integer *incx, real *beta,
|
|
real *y, integer *incy)
|
|
{
|
|
/* System generated locals */
|
|
integer a_dim1, a_offset, i__1, i__2;
|
|
real r__1;
|
|
|
|
/* Local variables */
|
|
integer info;
|
|
real temp;
|
|
integer lenx, leny;
|
|
extern integer ilatrans_(char *);
|
|
real safe1;
|
|
integer i__, j;
|
|
logical symb_zero__;
|
|
integer iy, jx, kx, ky;
|
|
extern real slamch_(char *);
|
|
extern /* Subroutine */ int xerbla_(char *, integer *, ftnlen);
|
|
|
|
|
|
/* -- LAPACK computational 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 */
|
|
|
|
|
|
/* ===================================================================== */
|
|
|
|
|
|
/* Test the input parameters. */
|
|
|
|
/* Parameter adjustments */
|
|
a_dim1 = *lda;
|
|
a_offset = 1 + a_dim1 * 1;
|
|
a -= a_offset;
|
|
--x;
|
|
--y;
|
|
|
|
/* Function Body */
|
|
info = 0;
|
|
if (! (*trans == ilatrans_("N") || *trans == ilatrans_("T") || *trans == ilatrans_("C"))) {
|
|
info = 1;
|
|
} else if (*m < 0) {
|
|
info = 2;
|
|
} else if (*n < 0) {
|
|
info = 3;
|
|
} else if (*lda < f2cmax(1,*m)) {
|
|
info = 6;
|
|
} else if (*incx == 0) {
|
|
info = 8;
|
|
} else if (*incy == 0) {
|
|
info = 11;
|
|
}
|
|
if (info != 0) {
|
|
xerbla_("SLA_GEAMV ", &info, (ftnlen)10);
|
|
return;
|
|
}
|
|
|
|
/* Quick return if possible. */
|
|
|
|
if (*m == 0 || *n == 0 || *alpha == 0.f && *beta == 1.f) {
|
|
return;
|
|
}
|
|
|
|
/* Set LENX and LENY, the lengths of the vectors x and y, and set */
|
|
/* up the start points in X and Y. */
|
|
|
|
if (*trans == ilatrans_("N")) {
|
|
lenx = *n;
|
|
leny = *m;
|
|
} else {
|
|
lenx = *m;
|
|
leny = *n;
|
|
}
|
|
if (*incx > 0) {
|
|
kx = 1;
|
|
} else {
|
|
kx = 1 - (lenx - 1) * *incx;
|
|
}
|
|
if (*incy > 0) {
|
|
ky = 1;
|
|
} else {
|
|
ky = 1 - (leny - 1) * *incy;
|
|
}
|
|
|
|
/* Set SAFE1 essentially to be the underflow threshold times the */
|
|
/* number of additions in each row. */
|
|
|
|
safe1 = slamch_("Safe minimum");
|
|
safe1 = (*n + 1) * safe1;
|
|
|
|
/* Form y := alpha*abs(A)*abs(x) + beta*abs(y). */
|
|
|
|
/* The O(M*N) SYMB_ZERO tests could be replaced by O(N) queries to */
|
|
/* the inexact flag. Still doesn't help change the iteration order */
|
|
/* to per-column. */
|
|
|
|
iy = ky;
|
|
if (*incx == 1) {
|
|
if (*trans == ilatrans_("N")) {
|
|
i__1 = leny;
|
|
for (i__ = 1; i__ <= i__1; ++i__) {
|
|
if (*beta == 0.f) {
|
|
symb_zero__ = TRUE_;
|
|
y[iy] = 0.f;
|
|
} else if (y[iy] == 0.f) {
|
|
symb_zero__ = TRUE_;
|
|
} else {
|
|
symb_zero__ = FALSE_;
|
|
y[iy] = *beta * (r__1 = y[iy], abs(r__1));
|
|
}
|
|
if (*alpha != 0.f) {
|
|
i__2 = lenx;
|
|
for (j = 1; j <= i__2; ++j) {
|
|
temp = (r__1 = a[i__ + j * a_dim1], abs(r__1));
|
|
symb_zero__ = symb_zero__ && (x[j] == 0.f || temp ==
|
|
0.f);
|
|
y[iy] += *alpha * (r__1 = x[j], abs(r__1)) * temp;
|
|
}
|
|
}
|
|
if (! symb_zero__) {
|
|
y[iy] += r_sign(&safe1, &y[iy]);
|
|
}
|
|
iy += *incy;
|
|
}
|
|
} else {
|
|
i__1 = leny;
|
|
for (i__ = 1; i__ <= i__1; ++i__) {
|
|
if (*beta == 0.f) {
|
|
symb_zero__ = TRUE_;
|
|
y[iy] = 0.f;
|
|
} else if (y[iy] == 0.f) {
|
|
symb_zero__ = TRUE_;
|
|
} else {
|
|
symb_zero__ = FALSE_;
|
|
y[iy] = *beta * (r__1 = y[iy], abs(r__1));
|
|
}
|
|
if (*alpha != 0.f) {
|
|
i__2 = lenx;
|
|
for (j = 1; j <= i__2; ++j) {
|
|
temp = (r__1 = a[j + i__ * a_dim1], abs(r__1));
|
|
symb_zero__ = symb_zero__ && (x[j] == 0.f || temp ==
|
|
0.f);
|
|
y[iy] += *alpha * (r__1 = x[j], abs(r__1)) * temp;
|
|
}
|
|
}
|
|
if (! symb_zero__) {
|
|
y[iy] += r_sign(&safe1, &y[iy]);
|
|
}
|
|
iy += *incy;
|
|
}
|
|
}
|
|
} else {
|
|
if (*trans == ilatrans_("N")) {
|
|
i__1 = leny;
|
|
for (i__ = 1; i__ <= i__1; ++i__) {
|
|
if (*beta == 0.f) {
|
|
symb_zero__ = TRUE_;
|
|
y[iy] = 0.f;
|
|
} else if (y[iy] == 0.f) {
|
|
symb_zero__ = TRUE_;
|
|
} else {
|
|
symb_zero__ = FALSE_;
|
|
y[iy] = *beta * (r__1 = y[iy], abs(r__1));
|
|
}
|
|
if (*alpha != 0.f) {
|
|
jx = kx;
|
|
i__2 = lenx;
|
|
for (j = 1; j <= i__2; ++j) {
|
|
temp = (r__1 = a[i__ + j * a_dim1], abs(r__1));
|
|
symb_zero__ = symb_zero__ && (x[jx] == 0.f || temp ==
|
|
0.f);
|
|
y[iy] += *alpha * (r__1 = x[jx], abs(r__1)) * temp;
|
|
jx += *incx;
|
|
}
|
|
}
|
|
if (! symb_zero__) {
|
|
y[iy] += r_sign(&safe1, &y[iy]);
|
|
}
|
|
iy += *incy;
|
|
}
|
|
} else {
|
|
i__1 = leny;
|
|
for (i__ = 1; i__ <= i__1; ++i__) {
|
|
if (*beta == 0.f) {
|
|
symb_zero__ = TRUE_;
|
|
y[iy] = 0.f;
|
|
} else if (y[iy] == 0.f) {
|
|
symb_zero__ = TRUE_;
|
|
} else {
|
|
symb_zero__ = FALSE_;
|
|
y[iy] = *beta * (r__1 = y[iy], abs(r__1));
|
|
}
|
|
if (*alpha != 0.f) {
|
|
jx = kx;
|
|
i__2 = lenx;
|
|
for (j = 1; j <= i__2; ++j) {
|
|
temp = (r__1 = a[j + i__ * a_dim1], abs(r__1));
|
|
symb_zero__ = symb_zero__ && (x[jx] == 0.f || temp ==
|
|
0.f);
|
|
y[iy] += *alpha * (r__1 = x[jx], abs(r__1)) * temp;
|
|
jx += *incx;
|
|
}
|
|
}
|
|
if (! symb_zero__) {
|
|
y[iy] += r_sign(&safe1, &y[iy]);
|
|
}
|
|
iy += *incy;
|
|
}
|
|
}
|
|
}
|
|
|
|
return;
|
|
|
|
/* End of SLA_GEAMV */
|
|
|
|
} /* sla_geamv__ */
|
|
|