|
1
2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 |
SUBROUTINE SGEQP3( M, N, A, LDA, JPVT, TAU, WORK, LWORK, INFO )
* * -- LAPACK routine (version 3.3.1) -- * -- LAPACK is a software package provided by Univ. of Tennessee, -- * -- Univ. of California Berkeley, Univ. of Colorado Denver and NAG Ltd..-- * -- April 2011 -- * * .. Scalar Arguments .. INTEGER INFO, LDA, LWORK, M, N * .. * .. Array Arguments .. INTEGER JPVT( * ) REAL A( LDA, * ), TAU( * ), WORK( * ) * .. * * Purpose * ======= * * SGEQP3 computes a QR factorization with column pivoting of a * matrix A: A*P = Q*R using Level 3 BLAS. * * Arguments * ========= * * M (input) INTEGER * The number of rows of the matrix A. M >= 0. * * N (input) INTEGER * The number of columns of the matrix A. N >= 0. * * A (input/output) REAL array, dimension (LDA,N) * On entry, the M-by-N matrix A. * On exit, the upper triangle of the array contains the * min(M,N)-by-N upper trapezoidal matrix R; the elements below * the diagonal, together with the array TAU, represent the * orthogonal matrix Q as a product of min(M,N) elementary * reflectors. * * LDA (input) INTEGER * The leading dimension of the array A. LDA >= max(1,M). * * JPVT (input/output) INTEGER array, dimension (N) * On entry, if JPVT(J).ne.0, the J-th column of A is permuted * to the front of A*P (a leading column); if JPVT(J)=0, * the J-th column of A is a free column. * On exit, if JPVT(J)=K, then the J-th column of A*P was the * the K-th column of A. * * TAU (output) REAL array, dimension (min(M,N)) * The scalar factors of the elementary reflectors. * * WORK (workspace/output) REAL array, dimension (MAX(1,LWORK)) * On exit, if INFO=0, WORK(1) returns the optimal LWORK. * * LWORK (input) INTEGER * The dimension of the array WORK. LWORK >= 3*N+1. * For optimal performance LWORK >= 2*N+( N+1 )*NB, where NB * is the optimal blocksize. * * 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. * * INFO (output) INTEGER * = 0: successful exit. * < 0: if INFO = -i, the i-th argument had an illegal value. * * Further Details * =============== * * The matrix Q is represented as a product of elementary reflectors * * Q = H(1) H(2) . . . H(k), where k = min(m,n). * * Each H(i) has the form * * H(i) = I - tau * v * v**T * * where tau is a real/complex scalar, and v is a real/complex vector * with v(1:i-1) = 0 and v(i) = 1; v(i+1:m) is stored on exit in * A(i+1:m,i), and tau in TAU(i). * * Based on contributions by * G. Quintana-Orti, Depto. de Informatica, Universidad Jaime I, Spain * X. Sun, Computer Science Dept., Duke University, USA * * ===================================================================== * * .. Parameters .. INTEGER INB, INBMIN, IXOVER PARAMETER ( INB = 1, INBMIN = 2, IXOVER = 3 ) * .. * .. Local Scalars .. LOGICAL LQUERY INTEGER FJB, IWS, J, JB, LWKOPT, MINMN, MINWS, NA, NB, $ NBMIN, NFXD, NX, SM, SMINMN, SN, TOPBMN * .. * .. External Subroutines .. EXTERNAL SGEQRF, SLAQP2, SLAQPS, SORMQR, SSWAP, XERBLA * .. * .. External Functions .. INTEGER ILAENV REAL SNRM2 EXTERNAL ILAENV, SNRM2 * .. * .. Intrinsic Functions .. INTRINSIC INT, MAX, MIN * .. * .. Executable Statements .. * INFO = 0 LQUERY = ( LWORK.EQ.-1 ) IF( M.LT.0 ) THEN INFO = -1 ELSE IF( N.LT.0 ) THEN INFO = -2 ELSE IF( LDA.LT.MAX( 1, M ) ) THEN INFO = -4 END IF * IF( INFO.EQ.0 ) THEN MINMN = MIN( M, N ) IF( MINMN.EQ.0 ) THEN IWS = 1 LWKOPT = 1 ELSE IWS = 3*N + 1 NB = ILAENV( INB, 'SGEQRF', ' ', M, N, -1, -1 ) LWKOPT = 2*N + ( N + 1 )*NB END IF WORK( 1 ) = LWKOPT * IF( ( LWORK.LT.IWS ) .AND. .NOT.LQUERY ) THEN INFO = -8 END IF END IF * IF( INFO.NE.0 ) THEN CALL XERBLA( 'SGEQP3', -INFO ) RETURN ELSE IF( LQUERY ) THEN RETURN END IF * * Quick return if possible. * IF( MINMN.EQ.0 ) THEN RETURN END IF * * Move initial columns up front. * NFXD = 1 DO 10 J = 1, N IF( JPVT( J ).NE.0 ) THEN IF( J.NE.NFXD ) THEN CALL SSWAP( M, A( 1, J ), 1, A( 1, NFXD ), 1 ) JPVT( J ) = JPVT( NFXD ) JPVT( NFXD ) = J ELSE JPVT( J ) = J END IF NFXD = NFXD + 1 ELSE JPVT( J ) = J END IF 10 CONTINUE NFXD = NFXD - 1 * * Factorize fixed columns * ======================= * * Compute the QR factorization of fixed columns and update * remaining columns. * IF( NFXD.GT.0 ) THEN NA = MIN( M, NFXD ) *CC CALL SGEQR2( M, NA, A, LDA, TAU, WORK, INFO ) CALL SGEQRF( M, NA, A, LDA, TAU, WORK, LWORK, INFO ) IWS = MAX( IWS, INT( WORK( 1 ) ) ) IF( NA.LT.N ) THEN *CC CALL SORM2R( 'Left', 'Transpose', M, N-NA, NA, A, LDA, *CC $ TAU, A( 1, NA+1 ), LDA, WORK, INFO ) CALL SORMQR( 'Left', 'Transpose', M, N-NA, NA, A, LDA, TAU, $ A( 1, NA+1 ), LDA, WORK, LWORK, INFO ) IWS = MAX( IWS, INT( WORK( 1 ) ) ) END IF END IF * * Factorize free columns * ====================== * IF( NFXD.LT.MINMN ) THEN * SM = M - NFXD SN = N - NFXD SMINMN = MINMN - NFXD * * Determine the block size. * NB = ILAENV( INB, 'SGEQRF', ' ', SM, SN, -1, -1 ) NBMIN = 2 NX = 0 * IF( ( NB.GT.1 ) .AND. ( NB.LT.SMINMN ) ) THEN * * Determine when to cross over from blocked to unblocked code. * NX = MAX( 0, ILAENV( IXOVER, 'SGEQRF', ' ', SM, SN, -1, $ -1 ) ) * * IF( NX.LT.SMINMN ) THEN * * Determine if workspace is large enough for blocked code. * MINWS = 2*SN + ( SN+1 )*NB IWS = MAX( IWS, MINWS ) IF( LWORK.LT.MINWS ) THEN * * Not enough workspace to use optimal NB: Reduce NB and * determine the minimum value of NB. * NB = ( LWORK-2*SN ) / ( SN+1 ) NBMIN = MAX( 2, ILAENV( INBMIN, 'SGEQRF', ' ', SM, SN, $ -1, -1 ) ) * * END IF END IF END IF * * Initialize partial column norms. The first N elements of work * store the exact column norms. * DO 20 J = NFXD + 1, N WORK( J ) = SNRM2( SM, A( NFXD+1, J ), 1 ) WORK( N+J ) = WORK( J ) 20 CONTINUE * IF( ( NB.GE.NBMIN ) .AND. ( NB.LT.SMINMN ) .AND. $ ( NX.LT.SMINMN ) ) THEN * * Use blocked code initially. * J = NFXD + 1 * * Compute factorization: while loop. * * TOPBMN = MINMN - NX 30 CONTINUE IF( J.LE.TOPBMN ) THEN JB = MIN( NB, TOPBMN-J+1 ) * * Factorize JB columns among columns J:N. * CALL SLAQPS( M, N-J+1, J-1, JB, FJB, A( 1, J ), LDA, $ JPVT( J ), TAU( J ), WORK( J ), WORK( N+J ), $ WORK( 2*N+1 ), WORK( 2*N+JB+1 ), N-J+1 ) * J = J + FJB GO TO 30 END IF ELSE J = NFXD + 1 END IF * * Use unblocked code to factor the last or only block. * * IF( J.LE.MINMN ) $ CALL SLAQP2( M, N-J+1, J-1, A( 1, J ), LDA, JPVT( J ), $ TAU( J ), WORK( J ), WORK( N+J ), $ WORK( 2*N+1 ) ) * END IF * WORK( 1 ) = IWS RETURN * * End of SGEQP3 * END |