1*> \brief \b DGEQRF 2* 3* =========== DOCUMENTATION =========== 4* 5* Online html documentation available at 6* http://www.netlib.org/lapack/explore-html/ 7* 8*> \htmlonly 9*> Download DGEQRF + dependencies 10*> <a href="http://www.netlib.org/cgi-bin/netlibfiles.tgz?format=tgz&filename=/lapack/lapack_routine/dgeqrf.f"> 11*> [TGZ]</a> 12*> <a href="http://www.netlib.org/cgi-bin/netlibfiles.zip?format=zip&filename=/lapack/lapack_routine/dgeqrf.f"> 13*> [ZIP]</a> 14*> <a href="http://www.netlib.org/cgi-bin/netlibfiles.txt?format=txt&filename=/lapack/lapack_routine/dgeqrf.f"> 15*> [TXT]</a> 16*> \endhtmlonly 17* 18* Definition: 19* =========== 20* 21* SUBROUTINE DGEQRF( M, N, A, LDA, TAU, WORK, LWORK, INFO ) 22* 23* .. Scalar Arguments .. 24* INTEGER INFO, LDA, LWORK, M, N 25* .. 26* .. Array Arguments .. 27* DOUBLE PRECISION A( LDA, * ), TAU( * ), WORK( * ) 28* .. 29* 30* 31*> \par Purpose: 32* ============= 33*> 34*> \verbatim 35*> 36*> DGEQRF computes a QR factorization of a real M-by-N matrix A: 37*> 38*> A = Q * ( R ), 39*> ( 0 ) 40*> 41*> where: 42*> 43*> Q is a M-by-M orthogonal matrix; 44*> R is an upper-triangular N-by-N matrix; 45*> 0 is a (M-N)-by-N zero matrix, if M > N. 46*> 47*> \endverbatim 48* 49* Arguments: 50* ========== 51* 52*> \param[in] M 53*> \verbatim 54*> M is INTEGER 55*> The number of rows of the matrix A. M >= 0. 56*> \endverbatim 57*> 58*> \param[in] N 59*> \verbatim 60*> N is INTEGER 61*> The number of columns of the matrix A. N >= 0. 62*> \endverbatim 63*> 64*> \param[in,out] A 65*> \verbatim 66*> A is DOUBLE PRECISION array, dimension (LDA,N) 67*> On entry, the M-by-N matrix A. 68*> On exit, the elements on and above the diagonal of the array 69*> contain the min(M,N)-by-N upper trapezoidal matrix R (R is 70*> upper triangular if m >= n); the elements below the diagonal, 71*> with the array TAU, represent the orthogonal matrix Q as a 72*> product of min(m,n) elementary reflectors (see Further 73*> Details). 74*> \endverbatim 75*> 76*> \param[in] LDA 77*> \verbatim 78*> LDA is INTEGER 79*> The leading dimension of the array A. LDA >= max(1,M). 80*> \endverbatim 81*> 82*> \param[out] TAU 83*> \verbatim 84*> TAU is DOUBLE PRECISION array, dimension (min(M,N)) 85*> The scalar factors of the elementary reflectors (see Further 86*> Details). 87*> \endverbatim 88*> 89*> \param[out] WORK 90*> \verbatim 91*> WORK is DOUBLE PRECISION array, dimension (MAX(1,LWORK)) 92*> On exit, if INFO = 0, WORK(1) returns the optimal LWORK. 93*> \endverbatim 94*> 95*> \param[in] LWORK 96*> \verbatim 97*> LWORK is INTEGER 98*> The dimension of the array WORK. LWORK >= max(1,N). 99*> For optimum performance LWORK >= N*NB, where NB is 100*> the optimal blocksize. 101*> 102*> If LWORK = -1, then a workspace query is assumed; the routine 103*> only calculates the optimal size of the WORK array, returns 104*> this value as the first entry of the WORK array, and no error 105*> message related to LWORK is issued by XERBLA. 106*> \endverbatim 107*> 108*> \param[out] INFO 109*> \verbatim 110*> INFO is INTEGER 111*> = 0: successful exit 112*> < 0: if INFO = -i, the i-th argument had an illegal value 113*> \endverbatim 114* 115* Authors: 116* ======== 117* 118*> \author Univ. of Tennessee 119*> \author Univ. of California Berkeley 120*> \author Univ. of Colorado Denver 121*> \author NAG Ltd. 122* 123*> \ingroup doubleGEcomputational 124* 125*> \par Further Details: 126* ===================== 127*> 128*> \verbatim 129*> 130*> The matrix Q is represented as a product of elementary reflectors 131*> 132*> Q = H(1) H(2) . . . H(k), where k = min(m,n). 133*> 134*> Each H(i) has the form 135*> 136*> H(i) = I - tau * v * v**T 137*> 138*> where tau is a real scalar, and v is a real vector with 139*> v(1:i-1) = 0 and v(i) = 1; v(i+1:m) is stored on exit in A(i+1:m,i), 140*> and tau in TAU(i). 141*> \endverbatim 142*> 143* ===================================================================== 144 SUBROUTINE DGEQRF( M, N, A, LDA, TAU, WORK, LWORK, INFO ) 145* 146* -- LAPACK computational routine -- 147* -- LAPACK is a software package provided by Univ. of Tennessee, -- 148* -- Univ. of California Berkeley, Univ. of Colorado Denver and NAG Ltd..-- 149* 150* .. Scalar Arguments .. 151 INTEGER INFO, LDA, LWORK, M, N 152* .. 153* .. Array Arguments .. 154 DOUBLE PRECISION A( LDA, * ), TAU( * ), WORK( * ) 155* .. 156* 157* ===================================================================== 158* 159* .. Local Scalars .. 160 LOGICAL LQUERY 161 INTEGER I, IB, IINFO, IWS, K, LDWORK, LWKOPT, NB, 162 $ NBMIN, NX 163* .. 164* .. External Subroutines .. 165 EXTERNAL DGEQR2, DLARFB, DLARFT, XERBLA 166* .. 167* .. Intrinsic Functions .. 168 INTRINSIC MAX, MIN 169* .. 170* .. External Functions .. 171 INTEGER ILAENV 172 EXTERNAL ILAENV 173* .. 174* .. Executable Statements .. 175* 176* Test the input arguments 177* 178 INFO = 0 179 NB = ILAENV( 1, 'DGEQRF', ' ', M, N, -1, -1 ) 180 LWKOPT = N*NB 181 WORK( 1 ) = LWKOPT 182 LQUERY = ( LWORK.EQ.-1 ) 183 IF( M.LT.0 ) THEN 184 INFO = -1 185 ELSE IF( N.LT.0 ) THEN 186 INFO = -2 187 ELSE IF( LDA.LT.MAX( 1, M ) ) THEN 188 INFO = -4 189 ELSE IF( LWORK.LT.MAX( 1, N ) .AND. .NOT.LQUERY ) THEN 190 INFO = -7 191 END IF 192 IF( INFO.NE.0 ) THEN 193 CALL XERBLA( 'DGEQRF', -INFO ) 194 RETURN 195 ELSE IF( LQUERY ) THEN 196 RETURN 197 END IF 198* 199* Quick return if possible 200* 201 K = MIN( M, N ) 202 IF( K.EQ.0 ) THEN 203 WORK( 1 ) = 1 204 RETURN 205 END IF 206* 207 NBMIN = 2 208 NX = 0 209 IWS = N 210 IF( NB.GT.1 .AND. NB.LT.K ) THEN 211* 212* Determine when to cross over from blocked to unblocked code. 213* 214 NX = MAX( 0, ILAENV( 3, 'DGEQRF', ' ', M, N, -1, -1 ) ) 215 IF( NX.LT.K ) THEN 216* 217* Determine if workspace is large enough for blocked code. 218* 219 LDWORK = N 220 IWS = LDWORK*NB 221 IF( LWORK.LT.IWS ) THEN 222* 223* Not enough workspace to use optimal NB: reduce NB and 224* determine the minimum value of NB. 225* 226 NB = LWORK / LDWORK 227 NBMIN = MAX( 2, ILAENV( 2, 'DGEQRF', ' ', M, N, -1, 228 $ -1 ) ) 229 END IF 230 END IF 231 END IF 232* 233 IF( NB.GE.NBMIN .AND. NB.LT.K .AND. NX.LT.K ) THEN 234* 235* Use blocked code initially 236* 237 DO 10 I = 1, K - NX, NB 238 IB = MIN( K-I+1, NB ) 239* 240* Compute the QR factorization of the current block 241* A(i:m,i:i+ib-1) 242* 243 CALL DGEQR2( M-I+1, IB, A( I, I ), LDA, TAU( I ), WORK, 244 $ IINFO ) 245 IF( I+IB.LE.N ) THEN 246* 247* Form the triangular factor of the block reflector 248* H = H(i) H(i+1) . . . H(i+ib-1) 249* 250 CALL DLARFT( 'Forward', 'Columnwise', M-I+1, IB, 251 $ A( I, I ), LDA, TAU( I ), WORK, LDWORK ) 252* 253* Apply H**T to A(i:m,i+ib:n) from the left 254* 255 CALL DLARFB( 'Left', 'Transpose', 'Forward', 256 $ 'Columnwise', M-I+1, N-I-IB+1, IB, 257 $ A( I, I ), LDA, WORK, LDWORK, A( I, I+IB ), 258 $ LDA, WORK( IB+1 ), LDWORK ) 259 END IF 260 10 CONTINUE 261 ELSE 262 I = 1 263 END IF 264* 265* Use unblocked code to factor the last or only block. 266* 267 IF( I.LE.K ) 268 $ CALL DGEQR2( M-I+1, N-I+1, A( I, I ), LDA, TAU( I ), WORK, 269 $ IINFO ) 270* 271 WORK( 1 ) = IWS 272 RETURN 273* 274* End of DGEQRF 275* 276 END 277