3 * =========== DOCUMENTATION ===========
5 * Online html documentation available at
6 * http://www.netlib.org/lapack/explore-html/
9 *> Download DTPQRT + dependencies
10 *> <a href="http://www.netlib.org/cgi-bin/netlibfiles.tgz?format=tgz&filename=/lapack/lapack_routine/dtpqrt.f">
12 *> <a href="http://www.netlib.org/cgi-bin/netlibfiles.zip?format=zip&filename=/lapack/lapack_routine/dtpqrt.f">
14 *> <a href="http://www.netlib.org/cgi-bin/netlibfiles.txt?format=txt&filename=/lapack/lapack_routine/dtpqrt.f">
21 * SUBROUTINE DTPQRT( M, N, L, NB, A, LDA, B, LDB, T, LDT, WORK,
24 * .. Scalar Arguments ..
25 * INTEGER INFO, LDA, LDB, LDT, N, M, L, NB
27 * .. Array Arguments ..
28 * DOUBLE PRECISION A( LDA, * ), B( LDB, * ), T( LDT, * ), WORK( * )
37 *> DTPQRT computes a blocked QR factorization of a real
38 *> "triangular-pentagonal" matrix C, which is composed of a
39 *> triangular block A and pentagonal block B, using the compact
40 *> WY representation for Q.
49 *> The number of rows of the matrix B.
56 *> The number of columns of the matrix B, and the order of the
57 *> triangular matrix A.
64 *> The number of rows of the upper trapezoidal part of B.
65 *> MIN(M,N) >= L >= 0. See Further Details.
71 *> The block size to be used in the blocked QR. N >= NB >= 1.
76 *> A is DOUBLE PRECISION array, dimension (LDA,N)
77 *> On entry, the upper triangular N-by-N matrix A.
78 *> On exit, the elements on and above the diagonal of the array
79 *> contain the upper triangular matrix R.
85 *> The leading dimension of the array A. LDA >= max(1,N).
90 *> B is DOUBLE PRECISION array, dimension (LDB,N)
91 *> On entry, the pentagonal M-by-N matrix B. The first M-L rows
92 *> are rectangular, and the last L rows are upper trapezoidal.
93 *> On exit, B contains the pentagonal matrix V. See Further Details.
99 *> The leading dimension of the array B. LDB >= max(1,M).
104 *> T is DOUBLE PRECISION array, dimension (LDT,N)
105 *> The upper triangular block reflectors stored in compact form
106 *> as a sequence of upper triangular blocks. See Further Details.
112 *> The leading dimension of the array T. LDT >= NB.
117 *> WORK is DOUBLE PRECISION array, dimension (NB*N)
123 *> = 0: successful exit
124 *> < 0: if INFO = -i, the i-th argument had an illegal value
130 *> \author Univ. of Tennessee
131 *> \author Univ. of California Berkeley
132 *> \author Univ. of Colorado Denver
135 *> \date November 2013
137 *> \ingroup doubleOTHERcomputational
139 *> \par Further Details:
140 * =====================
144 *> The input matrix C is a (N+M)-by-N matrix
149 *> where A is an upper triangular N-by-N matrix, and B is M-by-N pentagonal
150 *> matrix consisting of a (M-L)-by-N rectangular matrix B1 on top of a L-by-N
151 *> upper trapezoidal matrix B2:
153 *> B = [ B1 ] <- (M-L)-by-N rectangular
154 *> [ B2 ] <- L-by-N upper trapezoidal.
156 *> The upper trapezoidal matrix B2 consists of the first L rows of a
157 *> N-by-N upper triangular matrix, where 0 <= L <= MIN(M,N). If L=0,
158 *> B is rectangular M-by-N; if M=L=N, B is upper triangular.
160 *> The matrix W stores the elementary reflectors H(i) in the i-th column
161 *> below the diagonal (of A) in the (N+M)-by-N input matrix C
163 *> C = [ A ] <- upper triangular N-by-N
164 *> [ B ] <- M-by-N pentagonal
166 *> so that W can be represented as
168 *> W = [ I ] <- identity, N-by-N
169 *> [ V ] <- M-by-N, same form as B.
171 *> Thus, all of information needed for W is contained on exit in B, which
172 *> we call V above. Note that V has the same form as B; that is,
174 *> V = [ V1 ] <- (M-L)-by-N rectangular
175 *> [ V2 ] <- L-by-N upper trapezoidal.
177 *> The columns of V represent the vectors which define the H(i)'s.
179 *> The number of blocks is B = ceiling(N/NB), where each
180 *> block is of order NB except for the last block, which is of order
181 *> IB = N - (B-1)*NB. For each of the B blocks, a upper triangular block
182 *> reflector factor is computed: T1, T2, ..., TB. The NB-by-NB (and IB-by-IB
183 *> for the last block) T's are stored in the NB-by-N matrix T as
185 *> T = [T1 T2 ... TB].
188 * =====================================================================
189 SUBROUTINE DTPQRT( M, N, L, NB, A, LDA, B, LDB, T, LDT, WORK,
192 * -- LAPACK computational routine (version 3.5.0) --
193 * -- LAPACK is a software package provided by Univ. of Tennessee, --
194 * -- Univ. of California Berkeley, Univ. of Colorado Denver and NAG Ltd..--
197 * .. Scalar Arguments ..
198 INTEGER INFO, LDA, LDB, LDT, N, M, L, NB
200 * .. Array Arguments ..
201 DOUBLE PRECISION A( LDA, * ), B( LDB, * ), T( LDT, * ), WORK( * )
204 * =====================================================================
207 * .. Local Scalars ..
208 INTEGER I, IB, LB, MB, IINFO
210 * .. External Subroutines ..
211 EXTERNAL DTPQRT2, DTPRFB, XERBLA
213 * .. Executable Statements ..
215 * Test the input arguments
220 ELSE IF( N.LT.0 ) THEN
222 ELSE IF( L.LT.0 .OR. (L.GT.MIN(M,N) .AND. MIN(M,N).GE.0)) THEN
224 ELSE IF( NB.LT.1 .OR. (NB.GT.N .AND. N.GT.0)) THEN
226 ELSE IF( LDA.LT.MAX( 1, N ) ) THEN
228 ELSE IF( LDB.LT.MAX( 1, M ) ) THEN
230 ELSE IF( LDT.LT.NB ) THEN
234 CALL XERBLA( 'DTPQRT', -INFO )
238 * Quick return if possible
240 IF( M.EQ.0 .OR. N.EQ.0 ) RETURN
244 * Compute the QR factorization of the current block
246 IB = MIN( N-I+1, NB )
247 MB = MIN( M-L+I+IB-1, M )
254 CALL DTPQRT2( MB, IB, LB, A(I,I), LDA, B( 1, I ), LDB,
255 $ T(1, I ), LDT, IINFO )
257 * Update by applying H**T to B(:,I+IB:N) from the left
260 CALL DTPRFB( 'L', 'T', 'F', 'C', MB, N-I-IB+1, IB, LB,
261 $ B( 1, I ), LDB, T( 1, I ), LDT,
262 $ A( I, I+IB ), LDA, B( 1, I+IB ), LDB,