*> \brief \b ZGRQTS * * =========== DOCUMENTATION =========== * * Online html documentation available at * http://www.netlib.org/lapack/explore-html/ * * Definition: * =========== * * SUBROUTINE ZGRQTS( M, P, N, A, AF, Q, R, LDA, TAUA, B, BF, Z, T, * BWK, LDB, TAUB, WORK, LWORK, RWORK, RESULT ) * * .. Scalar Arguments .. * INTEGER LDA, LDB, LWORK, M, N, P * .. * .. Array Arguments .. * DOUBLE PRECISION RESULT( 4 ), RWORK( * ) * COMPLEX*16 A( LDA, * ), AF( LDA, * ), B( LDB, * ), * $ BF( LDB, * ), BWK( LDB, * ), Q( LDA, * ), * $ R( LDA, * ), T( LDB, * ), TAUA( * ), TAUB( * ), * $ WORK( LWORK ), Z( LDB, * ) * .. * * *> \par Purpose: * ============= *> *> \verbatim *> *> ZGRQTS tests ZGGRQF, which computes the GRQ factorization of an *> M-by-N matrix A and a P-by-N matrix B: A = R*Q and B = Z*T*Q. *> \endverbatim * * Arguments: * ========== * *> \param[in] M *> \verbatim *> M is INTEGER *> The number of rows of the matrix A. M >= 0. *> \endverbatim *> *> \param[in] P *> \verbatim *> P is INTEGER *> The number of rows of the matrix B. P >= 0. *> \endverbatim *> *> \param[in] N *> \verbatim *> N is INTEGER *> The number of columns of the matrices A and B. N >= 0. *> \endverbatim *> *> \param[in] A *> \verbatim *> A is COMPLEX*16 array, dimension (LDA,N) *> The M-by-N matrix A. *> \endverbatim *> *> \param[out] AF *> \verbatim *> AF is COMPLEX*16 array, dimension (LDA,N) *> Details of the GRQ factorization of A and B, as returned *> by ZGGRQF, see CGGRQF for further details. *> \endverbatim *> *> \param[out] Q *> \verbatim *> Q is COMPLEX*16 array, dimension (LDA,N) *> The N-by-N unitary matrix Q. *> \endverbatim *> *> \param[out] R *> \verbatim *> R is COMPLEX*16 array, dimension (LDA,MAX(M,N)) *> \endverbatim *> *> \param[in] LDA *> \verbatim *> LDA is INTEGER *> The leading dimension of the arrays A, AF, R and Q. *> LDA >= max(M,N). *> \endverbatim *> *> \param[out] TAUA *> \verbatim *> TAUA is COMPLEX*16 array, dimension (min(M,N)) *> The scalar factors of the elementary reflectors, as returned *> by DGGQRC. *> \endverbatim *> *> \param[in] B *> \verbatim *> B is COMPLEX*16 array, dimension (LDB,N) *> On entry, the P-by-N matrix A. *> \endverbatim *> *> \param[out] BF *> \verbatim *> BF is COMPLEX*16 array, dimension (LDB,N) *> Details of the GQR factorization of A and B, as returned *> by ZGGRQF, see CGGRQF for further details. *> \endverbatim *> *> \param[out] Z *> \verbatim *> Z is DOUBLE PRECISION array, dimension (LDB,P) *> The P-by-P unitary matrix Z. *> \endverbatim *> *> \param[out] T *> \verbatim *> T is COMPLEX*16 array, dimension (LDB,max(P,N)) *> \endverbatim *> *> \param[out] BWK *> \verbatim *> BWK is COMPLEX*16 array, dimension (LDB,N) *> \endverbatim *> *> \param[in] LDB *> \verbatim *> LDB is INTEGER *> The leading dimension of the arrays B, BF, Z and T. *> LDB >= max(P,N). *> \endverbatim *> *> \param[out] TAUB *> \verbatim *> TAUB is COMPLEX*16 array, dimension (min(P,N)) *> The scalar factors of the elementary reflectors, as returned *> by DGGRQF. *> \endverbatim *> *> \param[out] WORK *> \verbatim *> WORK is COMPLEX*16 array, dimension (LWORK) *> \endverbatim *> *> \param[in] LWORK *> \verbatim *> LWORK is INTEGER *> The dimension of the array WORK, LWORK >= max(M,P,N)**2. *> \endverbatim *> *> \param[out] RWORK *> \verbatim *> RWORK is DOUBLE PRECISION array, dimension (M) *> \endverbatim *> *> \param[out] RESULT *> \verbatim *> RESULT is DOUBLE PRECISION array, dimension (4) *> The test ratios: *> RESULT(1) = norm( R - A*Q' ) / ( MAX(M,N)*norm(A)*ULP) *> RESULT(2) = norm( T*Q - Z'*B ) / (MAX(P,N)*norm(B)*ULP) *> RESULT(3) = norm( I - Q'*Q ) / ( N*ULP ) *> RESULT(4) = norm( I - Z'*Z ) / ( P*ULP ) *> \endverbatim * * Authors: * ======== * *> \author Univ. of Tennessee *> \author Univ. of California Berkeley *> \author Univ. of Colorado Denver *> \author NAG Ltd. * *> \date November 2011 * *> \ingroup complex16_eig * * ===================================================================== SUBROUTINE ZGRQTS( M, P, N, A, AF, Q, R, LDA, TAUA, B, BF, Z, T, $ BWK, LDB, TAUB, WORK, LWORK, RWORK, RESULT ) * * -- LAPACK test routine (version 3.4.0) -- * -- LAPACK is a software package provided by Univ. of Tennessee, -- * -- Univ. of California Berkeley, Univ. of Colorado Denver and NAG Ltd..-- * November 2011 * * .. Scalar Arguments .. INTEGER LDA, LDB, LWORK, M, N, P * .. * .. Array Arguments .. DOUBLE PRECISION RESULT( 4 ), RWORK( * ) COMPLEX*16 A( LDA, * ), AF( LDA, * ), B( LDB, * ), $ BF( LDB, * ), BWK( LDB, * ), Q( LDA, * ), $ R( LDA, * ), T( LDB, * ), TAUA( * ), TAUB( * ), $ WORK( LWORK ), Z( LDB, * ) * .. * * ===================================================================== * * .. Parameters .. DOUBLE PRECISION ZERO, ONE PARAMETER ( ZERO = 0.0D+0, ONE = 1.0D+0 ) COMPLEX*16 CZERO, CONE PARAMETER ( CZERO = ( 0.0D+0, 0.0D+0 ), $ CONE = ( 1.0D+0, 0.0D+0 ) ) COMPLEX*16 CROGUE PARAMETER ( CROGUE = ( -1.0D+10, 0.0D+0 ) ) * .. * .. Local Scalars .. INTEGER INFO DOUBLE PRECISION ANORM, BNORM, RESID, ULP, UNFL * .. * .. External Functions .. DOUBLE PRECISION DLAMCH, ZLANGE, ZLANHE EXTERNAL DLAMCH, ZLANGE, ZLANHE * .. * .. External Subroutines .. EXTERNAL ZGEMM, ZGGRQF, ZHERK, ZLACPY, ZLASET, ZUNGQR, $ ZUNGRQ * .. * .. Intrinsic Functions .. INTRINSIC DBLE, MAX, MIN * .. * .. Executable Statements .. * ULP = DLAMCH( 'Precision' ) UNFL = DLAMCH( 'Safe minimum' ) * * Copy the matrix A to the array AF. * CALL ZLACPY( 'Full', M, N, A, LDA, AF, LDA ) CALL ZLACPY( 'Full', P, N, B, LDB, BF, LDB ) * ANORM = MAX( ZLANGE( '1', M, N, A, LDA, RWORK ), UNFL ) BNORM = MAX( ZLANGE( '1', P, N, B, LDB, RWORK ), UNFL ) * * Factorize the matrices A and B in the arrays AF and BF. * CALL ZGGRQF( M, P, N, AF, LDA, TAUA, BF, LDB, TAUB, WORK, LWORK, $ INFO ) * * Generate the N-by-N matrix Q * CALL ZLASET( 'Full', N, N, CROGUE, CROGUE, Q, LDA ) IF( M.LE.N ) THEN IF( M.GT.0 .AND. M.LT.N ) $ CALL ZLACPY( 'Full', M, N-M, AF, LDA, Q( N-M+1, 1 ), LDA ) IF( M.GT.1 ) $ CALL ZLACPY( 'Lower', M-1, M-1, AF( 2, N-M+1 ), LDA, $ Q( N-M+2, N-M+1 ), LDA ) ELSE IF( N.GT.1 ) $ CALL ZLACPY( 'Lower', N-1, N-1, AF( M-N+2, 1 ), LDA, $ Q( 2, 1 ), LDA ) END IF CALL ZUNGRQ( N, N, MIN( M, N ), Q, LDA, TAUA, WORK, LWORK, INFO ) * * Generate the P-by-P matrix Z * CALL ZLASET( 'Full', P, P, CROGUE, CROGUE, Z, LDB ) IF( P.GT.1 ) $ CALL ZLACPY( 'Lower', P-1, N, BF( 2, 1 ), LDB, Z( 2, 1 ), LDB ) CALL ZUNGQR( P, P, MIN( P, N ), Z, LDB, TAUB, WORK, LWORK, INFO ) * * Copy R * CALL ZLASET( 'Full', M, N, CZERO, CZERO, R, LDA ) IF( M.LE.N ) THEN CALL ZLACPY( 'Upper', M, M, AF( 1, N-M+1 ), LDA, R( 1, N-M+1 ), $ LDA ) ELSE CALL ZLACPY( 'Full', M-N, N, AF, LDA, R, LDA ) CALL ZLACPY( 'Upper', N, N, AF( M-N+1, 1 ), LDA, R( M-N+1, 1 ), $ LDA ) END IF * * Copy T * CALL ZLASET( 'Full', P, N, CZERO, CZERO, T, LDB ) CALL ZLACPY( 'Upper', P, N, BF, LDB, T, LDB ) * * Compute R - A*Q' * CALL ZGEMM( 'No transpose', 'Conjugate transpose', M, N, N, -CONE, $ A, LDA, Q, LDA, CONE, R, LDA ) * * Compute norm( R - A*Q' ) / ( MAX(M,N)*norm(A)*ULP ) . * RESID = ZLANGE( '1', M, N, R, LDA, RWORK ) IF( ANORM.GT.ZERO ) THEN RESULT( 1 ) = ( ( RESID / DBLE( MAX( 1, M, N ) ) ) / ANORM ) / $ ULP ELSE RESULT( 1 ) = ZERO END IF * * Compute T*Q - Z'*B * CALL ZGEMM( 'Conjugate transpose', 'No transpose', P, N, P, CONE, $ Z, LDB, B, LDB, CZERO, BWK, LDB ) CALL ZGEMM( 'No transpose', 'No transpose', P, N, N, CONE, T, LDB, $ Q, LDA, -CONE, BWK, LDB ) * * Compute norm( T*Q - Z'*B ) / ( MAX(P,N)*norm(A)*ULP ) . * RESID = ZLANGE( '1', P, N, BWK, LDB, RWORK ) IF( BNORM.GT.ZERO ) THEN RESULT( 2 ) = ( ( RESID / DBLE( MAX( 1, P, M ) ) ) / BNORM ) / $ ULP ELSE RESULT( 2 ) = ZERO END IF * * Compute I - Q*Q' * CALL ZLASET( 'Full', N, N, CZERO, CONE, R, LDA ) CALL ZHERK( 'Upper', 'No Transpose', N, N, -ONE, Q, LDA, ONE, R, $ LDA ) * * Compute norm( I - Q'*Q ) / ( N * ULP ) . * RESID = ZLANHE( '1', 'Upper', N, R, LDA, RWORK ) RESULT( 3 ) = ( RESID / DBLE( MAX( 1, N ) ) ) / ULP * * Compute I - Z'*Z * CALL ZLASET( 'Full', P, P, CZERO, CONE, T, LDB ) CALL ZHERK( 'Upper', 'Conjugate transpose', P, P, -ONE, Z, LDB, $ ONE, T, LDB ) * * Compute norm( I - Z'*Z ) / ( P*ULP ) . * RESID = ZLANHE( '1', 'Upper', P, T, LDB, RWORK ) RESULT( 4 ) = ( RESID / DBLE( MAX( 1, P ) ) ) / ULP * RETURN * * End of ZGRQTS * END