OpenBLAS/lapack-netlib/SRC/slamswlq.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)
*/




/* Table of constant values */

static integer c__0 = 0;

/* > \brief \b SLAMSWLQ */

/*  Definition: */
/*  =========== */

/*      SUBROUTINE SLAMSWLQ( SIDE, TRANS, M, N, K, MB, NB, A, LDA, T, */
/*     $                LDT, C, LDC, WORK, LWORK, INFO ) */


/*      CHARACTER         SIDE, TRANS */
/*      INTEGER           INFO, LDA, M, N, K, MB, NB, LDT, LWORK, LDC */
/*      DOUBLE        A( LDA, * ), WORK( * ), C(LDC, * ), */
/*     $                  T( LDT, * ) */
/* > \par Purpose: */
/*  ============= */
/* > */
/* > \verbatim */
/* > */
/* >    SLAMSWLQ overwrites the general real M-by-N matrix C with */
/* > */
/* > */
/* >                    SIDE = 'L'     SIDE = 'R' */
/* >    TRANS = 'N':      Q * C          C * Q */
/* >    TRANS = 'T':      Q**T * C       C * Q**T */
/* >    where Q is a real orthogonal matrix defined as the product of blocked */
/* >    elementary reflectors computed by short wide LQ */
/* >    factorization (SLASWLQ) */
/* > \endverbatim */

/*  Arguments: */
/*  ========== */

/* > \param[in] SIDE */
/* > \verbatim */
/* >          SIDE is CHARACTER*1 */
/* >          = 'L': apply Q or Q**T from the Left; */
/* >          = 'R': apply Q or Q**T from the Right. */
/* > \endverbatim */
/* > */
/* > \param[in] TRANS */
/* > \verbatim */
/* >          TRANS is CHARACTER*1 */
/* >          = 'N':  No transpose, apply Q; */
/* >          = 'T':  Transpose, apply Q**T. */
/* > \endverbatim */
/* > */
/* > \param[in] M */
/* > \verbatim */
/* >          M is INTEGER */
/* >          The number of rows of the matrix C.  M >=0. */
/* > \endverbatim */
/* > */
/* > \param[in] N */
/* > \verbatim */
/* >          N is INTEGER */
/* >          The number of columns of the matrix C. N >= M. */
/* > \endverbatim */
/* > */
/* > \param[in] K */
/* > \verbatim */
/* >          K is INTEGER */
/* >          The number of elementary reflectors whose product defines */
/* >          the matrix Q. */
/* >          M >= K >= 0; */
/* > */
/* > \endverbatim */
/* > \param[in] MB */
/* > \verbatim */
/* >          MB is INTEGER */
/* >          The row block size to be used in the blocked QR. */
/* >          M >= MB >= 1 */
/* > \endverbatim */
/* > */
/* > \param[in] NB */
/* > \verbatim */
/* >          NB is INTEGER */
/* >          The column block size to be used in the blocked QR. */
/* >          NB > M. */
/* > \endverbatim */
/* > */
/* > \param[in] NB */
/* > \verbatim */
/* >          NB is INTEGER */
/* >          The block size to be used in the blocked QR. */
/* >                MB > M. */
/* > */
/* > \endverbatim */
/* > */
/* > \param[in] A */
/* > \verbatim */
/* >          A is REAL array, dimension */
/* >                               (LDA,M) if SIDE = 'L', */
/* >                               (LDA,N) if SIDE = 'R' */
/* >          The i-th row must contain the vector which defines the blocked */
/* >          elementary reflector H(i), for i = 1,2,...,k, as returned by */
/* >          SLASWLQ in the first k rows of its array argument A. */
/* > \endverbatim */
/* > */
/* > \param[in] LDA */
/* > \verbatim */
/* >          LDA is INTEGER */
/* >          The leading dimension of the array A. */
/* >          If SIDE = 'L', LDA >= f2cmax(1,M); */
/* >          if SIDE = 'R', LDA >= f2cmax(1,N). */
/* > \endverbatim */
/* > */
/* > \param[in] T */
/* > \verbatim */
/* >          T is REAL array, dimension */
/* >          ( M * Number of blocks(CEIL(N-K/NB-K)), */
/* >          The blocked upper triangular block reflectors stored in compact form */
/* >          as a sequence of upper triangular blocks.  See below */
/* >          for further details. */
/* > \endverbatim */
/* > */
/* > \param[in] LDT */
/* > \verbatim */
/* >          LDT is INTEGER */
/* >          The leading dimension of the array T.  LDT >= MB. */
/* > \endverbatim */
/* > */
/* > \param[in,out] C */
/* > \verbatim */
/* >          C is REAL array, dimension (LDC,N) */
/* >          On entry, the M-by-N matrix C. */
/* >          On exit, C is overwritten by Q*C or Q**T*C or C*Q**T or C*Q. */
/* > \endverbatim */
/* > */
/* > \param[in] LDC */
/* > \verbatim */
/* >          LDC is INTEGER */
/* >          The leading dimension of the array C. LDC >= f2cmax(1,M). */
/* > \endverbatim */
/* > */
/* > \param[out] WORK */
/* > \verbatim */
/* >         (workspace) REAL array, dimension (MAX(1,LWORK)) */
/* > \endverbatim */
/* > */
/* > \param[in] LWORK */
/* > \verbatim */
/* >          LWORK is INTEGER */
/* >          The dimension of the array WORK. */
/* >          If SIDE = 'L', LWORK >= f2cmax(1,NB) * MB; */
/* >          if SIDE = 'R', LWORK >= f2cmax(1,M) * MB. */
/* >          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. */
/* > \endverbatim */
/* > */
/* > \param[out] INFO */
/* > \verbatim */
/* >          INFO is INTEGER */
/* >          = 0:  successful exit */
/* >          < 0:  if INFO = -i, the i-th argument had an illegal value */
/* > \endverbatim */

/*  Authors: */
/*  ======== */

/* > \author Univ. of Tennessee */
/* > \author Univ. of California Berkeley */
/* > \author Univ. of Colorado Denver */
/* > \author NAG Ltd. */

/* > \par Further Details: */
/*  ===================== */
/* > */
/* > \verbatim */
/* > Short-Wide LQ (SWLQ) performs LQ by a sequence of orthogonal transformations, */
/* > representing Q as a product of other orthogonal matrices */
/* >   Q = Q(1) * Q(2) * . . . * Q(k) */
/* > where each Q(i) zeros out upper diagonal entries of a block of NB rows of A: */
/* >   Q(1) zeros out the upper diagonal entries of rows 1:NB of A */
/* >   Q(2) zeros out the bottom MB-N rows of rows [1:M,NB+1:2*NB-M] of A */
/* >   Q(3) zeros out the bottom MB-N rows of rows [1:M,2*NB-M+1:3*NB-2*M] of A */
/* >   . . . */
/* > */
/* > Q(1) is computed by GELQT, which represents Q(1) by Householder vectors */
/* > stored under the diagonal of rows 1:MB of A, and by upper triangular */
/* > block reflectors, stored in array T(1:LDT,1:N). */
/* > For more information see Further Details in GELQT. */
/* > */
/* > Q(i) for i>1 is computed by TPLQT, which represents Q(i) by Householder vectors */
/* > stored in columns [(i-1)*(NB-M)+M+1:i*(NB-M)+M] of A, and by upper triangular */
/* > block reflectors, stored in array T(1:LDT,(i-1)*M+1:i*M). */
/* > The last Q(k) may use fewer rows. */
/* > For more information see Further Details in TPQRT. */
/* > */
/* > For more details of the overall algorithm, see the description of */
/* > Sequential TSQR in Section 2.2 of [1]. */
/* > */
/* > [1] “Communication-Optimal Parallel and Sequential QR and LU Factorizations, */
/* >     J. Demmel, L. Grigori, M. Hoemmen, J. Langou, */
/* >     SIAM J. Sci. Comput, vol. 34, no. 1, 2012 */
/* > \endverbatim */
/* > */
/*  ===================================================================== */
/* Subroutine */ int slamswlq_(char *side, char *trans, integer *m, integer *
	n, integer *k, integer *mb, integer *nb, real *a, integer *lda, real *
	t, integer *ldt, real *c__, integer *ldc, real *work, integer *lwork, 
	integer *info)
{
    /* System generated locals */
    integer a_dim1, a_offset, c_dim1, c_offset, t_dim1, t_offset, i__1, i__2, 
	    i__3;

    /* Local variables */
    logical left, tran;
    integer i__;
    extern logical lsame_(char *, char *);
    logical right;
    integer ii, kk, lw;
    extern /* Subroutine */ int xerbla_(char *, integer *, ftnlen);
    logical notran, lquery;
    integer ctr;
    extern /* Subroutine */ int sgemlqt_(char *, char *, integer *, integer *,
	     integer *, integer *, real *, integer *, real *, integer *, real 
	    *, integer *, real *, integer *), stpmlqt_(char *,
	     char *, integer *, integer *, integer *, integer *, integer *, 
	    real *, integer *, real *, integer *, real *, integer *, real *, 
	    integer *, real *, integer *);


/*  -- 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 arguments */

    /* Parameter adjustments */
    a_dim1 = *lda;
    a_offset = 1 + a_dim1 * 1;
    a -= a_offset;
    t_dim1 = *ldt;
    t_offset = 1 + t_dim1 * 1;
    t -= t_offset;
    c_dim1 = *ldc;
    c_offset = 1 + c_dim1 * 1;
    c__ -= c_offset;
    --work;

    /* Function Body */
    lquery = *lwork < 0;
    notran = lsame_(trans, "N");
    tran = lsame_(trans, "T");
    left = lsame_(side, "L");
    right = lsame_(side, "R");
    if (left) {
	lw = *n * *mb;
    } else {
	lw = *m * *mb;
    }

    *info = 0;
    if (! left && ! right) {
	*info = -1;
    } else if (! tran && ! notran) {
	*info = -2;
    } else if (*m < 0) {
	*info = -3;
    } else if (*n < 0) {
	*info = -4;
    } else if (*k < 0) {
	*info = -5;
    } else if (*lda < f2cmax(1,*k)) {
	*info = -9;
    } else if (*ldt < f2cmax(1,*mb)) {
	*info = -11;
    } else if (*ldc < f2cmax(1,*m)) {
	*info = -13;
    } else if (*lwork < f2cmax(1,lw) && ! lquery) {
	*info = -15;
    }

    if (*info != 0) {
	i__1 = -(*info);
	xerbla_("SLAMSWLQ", &i__1, (ftnlen)8);
	work[1] = (real) lw;
	return 0;
    } else if (lquery) {
	work[1] = (real) lw;
	return 0;
    }

/*     Quick return if possible */

/* Computing MIN */
    i__1 = f2cmin(*m,*n);
    if (f2cmin(i__1,*k) == 0) {
	return 0;
    }

/* Computing MAX */
    i__1 = f2cmax(*m,*n);
    if (*nb <= *k || *nb >= f2cmax(i__1,*k)) {
	sgemlqt_(side, trans, m, n, k, mb, &a[a_offset], lda, &t[t_offset], 
		ldt, &c__[c_offset], ldc, &work[1], info);
	return 0;
    }

    if (left && tran) {

/*         Multiply Q to the last block of C */

	kk = (*m - *k) % (*nb - *k);
	ctr = (*m - *k) / (*nb - *k);

	if (kk > 0) {
	    ii = *m - kk + 1;
	    stpmlqt_("L", "T", &kk, n, k, &c__0, mb, &a[ii * a_dim1 + 1], lda,
		     &t[(ctr * *k + 1) * t_dim1 + 1], ldt, &c__[c_dim1 + 1], 
		    ldc, &c__[ii + c_dim1], ldc, &work[1], info);
	} else {
	    ii = *m + 1;
	}

	i__1 = *nb + 1;
	i__2 = -(*nb - *k);
	for (i__ = ii - (*nb - *k); i__2 < 0 ? i__ >= i__1 : i__ <= i__1; i__ 
		+= i__2) {

/*         Multiply Q to the current block of C (1:M,I:I+NB) */

	    --ctr;
	    i__3 = *nb - *k;
	    stpmlqt_("L", "T", &i__3, n, k, &c__0, mb, &a[i__ * a_dim1 + 1], 
		    lda, &t[(ctr * *k + 1) * t_dim1 + 1], ldt, &c__[c_dim1 + 
		    1], ldc, &c__[i__ + c_dim1], ldc, &work[1], info);
	}

/*         Multiply Q to the first block of C (1:M,1:NB) */

	sgemlqt_("L", "T", nb, n, k, mb, &a[a_dim1 + 1], lda, &t[t_offset], 
		ldt, &c__[c_dim1 + 1], ldc, &work[1], info);

    } else if (left && notran) {

/*         Multiply Q to the first block of C */

	kk = (*m - *k) % (*nb - *k);
	ii = *m - kk + 1;
	ctr = 1;
	sgemlqt_("L", "N", nb, n, k, mb, &a[a_dim1 + 1], lda, &t[t_offset], 
		ldt, &c__[c_dim1 + 1], ldc, &work[1], info);

	i__2 = ii - *nb + *k;
	i__1 = *nb - *k;
	for (i__ = *nb + 1; i__1 < 0 ? i__ >= i__2 : i__ <= i__2; i__ += i__1)
		 {

/*         Multiply Q to the current block of C (I:I+NB,1:N) */

	    i__3 = *nb - *k;
	    stpmlqt_("L", "N", &i__3, n, k, &c__0, mb, &a[i__ * a_dim1 + 1], 
		    lda, &t[(ctr * *k + 1) * t_dim1 + 1], ldt, &c__[c_dim1 + 
		    1], ldc, &c__[i__ + c_dim1], ldc, &work[1], info);
	    ++ctr;

	}
	if (ii <= *m) {

/*         Multiply Q to the last block of C */

	    stpmlqt_("L", "N", &kk, n, k, &c__0, mb, &a[ii * a_dim1 + 1], lda,
		     &t[(ctr * *k + 1) * t_dim1 + 1], ldt, &c__[c_dim1 + 1], 
		    ldc, &c__[ii + c_dim1], ldc, &work[1], info);

	}

    } else if (right && notran) {

/*         Multiply Q to the last block of C */

	kk = (*n - *k) % (*nb - *k);
	ctr = (*n - *k) / (*nb - *k);
	if (kk > 0) {
	    ii = *n - kk + 1;
	    stpmlqt_("R", "N", m, &kk, k, &c__0, mb, &a[ii * a_dim1 + 1], lda,
		     &t[(ctr * *k + 1) * t_dim1 + 1], ldt, &c__[c_dim1 + 1], 
		    ldc, &c__[ii * c_dim1 + 1], ldc, &work[1], info);
	} else {
	    ii = *n + 1;
	}

	i__1 = *nb + 1;
	i__2 = -(*nb - *k);
	for (i__ = ii - (*nb - *k); i__2 < 0 ? i__ >= i__1 : i__ <= i__1; i__ 
		+= i__2) {

/*         Multiply Q to the current block of C (1:M,I:I+MB) */

	    --ctr;
	    i__3 = *nb - *k;
	    stpmlqt_("R", "N", m, &i__3, k, &c__0, mb, &a[i__ * a_dim1 + 1], 
		    lda, &t[(ctr * *k + 1) * t_dim1 + 1], ldt, &c__[c_dim1 + 
		    1], ldc, &c__[i__ * c_dim1 + 1], ldc, &work[1], info);
	}

/*         Multiply Q to the first block of C (1:M,1:MB) */

	sgemlqt_("R", "N", m, nb, k, mb, &a[a_dim1 + 1], lda, &t[t_offset], 
		ldt, &c__[c_dim1 + 1], ldc, &work[1], info);

    } else if (right && tran) {

/*       Multiply Q to the first block of C */

	kk = (*n - *k) % (*nb - *k);
	ii = *n - kk + 1;
	ctr = 1;
	sgemlqt_("R", "T", m, nb, k, mb, &a[a_dim1 + 1], lda, &t[t_offset], 
		ldt, &c__[c_dim1 + 1], ldc, &work[1], info);

	i__2 = ii - *nb + *k;
	i__1 = *nb - *k;
	for (i__ = *nb + 1; i__1 < 0 ? i__ >= i__2 : i__ <= i__2; i__ += i__1)
		 {

/*         Multiply Q to the current block of C (1:M,I:I+MB) */

	    i__3 = *nb - *k;
	    stpmlqt_("R", "T", m, &i__3, k, &c__0, mb, &a[i__ * a_dim1 + 1], 
		    lda, &t[(ctr * *k + 1) * t_dim1 + 1], ldt, &c__[c_dim1 + 
		    1], ldc, &c__[i__ * c_dim1 + 1], ldc, &work[1], info);
	    ++ctr;

	}
	if (ii <= *n) {

/*       Multiply Q to the last block of C */

	    stpmlqt_("R", "T", m, &kk, k, &c__0, mb, &a[ii * a_dim1 + 1], lda,
		     &t[(ctr * *k + 1) * t_dim1 + 1], ldt, &c__[c_dim1 + 1], 
		    ldc, &c__[ii * c_dim1 + 1], ldc, &work[1], info);

	}

    }

    work[1] = (real) lw;
    return 0;

/*     End of SLAMSWLQ */

} /* slamswlq_ */