*> \brief \b CGQRTS * * =========== DOCUMENTATION =========== * * Online html documentation available at * http://www.netlib.org/lapack/explore-html/ * * Definition: * =========== * * SUBROUTINE CGQRTS( N, M, P, A, AF, Q, R, LDA, TAUA, B, BF, Z, T, * BWK, LDB, TAUB, WORK, LWORK, RWORK, RESULT ) * * .. Scalar Arguments .. * INTEGER LDA, LDB, LWORK, M, P, N * .. * .. Array Arguments .. * REAL RWORK( * ), RESULT( 4 ) * COMPLEX A( LDA, * ), AF( LDA, * ), R( LDA, * ), * $ Q( LDA, * ), B( LDB, * ), BF( LDB, * ), * $ T( LDB, * ), Z( LDB, * ), BWK( LDB, * ), * $ TAUA( * ), TAUB( * ), WORK( LWORK ) * .. * * *> \par Purpose: * ============= *> *> \verbatim *> *> CGQRTS tests CGGQRF, which computes the GQR factorization of an *> N-by-M matrix A and a N-by-P matrix B: A = Q*R and B = Q*T*Z. *> \endverbatim * * Arguments: * ========== * *> \param[in] N *> \verbatim *> N is INTEGER *> The number of rows of the matrices A and B. N >= 0. *> \endverbatim *> *> \param[in] M *> \verbatim *> M is INTEGER *> The number of columns of the matrix A. M >= 0. *> \endverbatim *> *> \param[in] P *> \verbatim *> P is INTEGER *> The number of columns of the matrix B. P >= 0. *> \endverbatim *> *> \param[in] A *> \verbatim *> A is COMPLEX array, dimension (LDA,M) *> The N-by-M matrix A. *> \endverbatim *> *> \param[out] AF *> \verbatim *> AF is COMPLEX array, dimension (LDA,N) *> Details of the GQR factorization of A and B, as returned *> by CGGQRF, see CGGQRF for further details. *> \endverbatim *> *> \param[out] Q *> \verbatim *> Q is COMPLEX array, dimension (LDA,N) *> The M-by-M unitary matrix Q. *> \endverbatim *> *> \param[out] R *> \verbatim *> R is COMPLEX 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 array, dimension (min(M,N)) *> The scalar factors of the elementary reflectors, as returned *> by CGGQRF. *> \endverbatim *> *> \param[in] B *> \verbatim *> B is COMPLEX array, dimension (LDB,P) *> On entry, the N-by-P matrix A. *> \endverbatim *> *> \param[out] BF *> \verbatim *> BF is COMPLEX array, dimension (LDB,N) *> Details of the GQR factorization of A and B, as returned *> by CGGQRF, see CGGQRF for further details. *> \endverbatim *> *> \param[out] Z *> \verbatim *> Z is COMPLEX array, dimension (LDB,P) *> The P-by-P unitary matrix Z. *> \endverbatim *> *> \param[out] T *> \verbatim *> T is COMPLEX array, dimension (LDB,max(P,N)) *> \endverbatim *> *> \param[out] BWK *> \verbatim *> BWK is COMPLEX 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 array, dimension (min(P,N)) *> The scalar factors of the elementary reflectors, as returned *> by SGGRQF. *> \endverbatim *> *> \param[out] WORK *> \verbatim *> WORK is COMPLEX array, dimension (LWORK) *> \endverbatim *> *> \param[in] LWORK *> \verbatim *> LWORK is INTEGER *> The dimension of the array WORK, LWORK >= max(N,M,P)**2. *> \endverbatim *> *> \param[out] RWORK *> \verbatim *> RWORK is REAL array, dimension (max(N,M,P)) *> \endverbatim *> *> \param[out] RESULT *> \verbatim *> RESULT is REAL array, dimension (4) *> The test ratios: *> RESULT(1) = norm( R - Q'*A ) / ( MAX(M,N)*norm(A)*ULP) *> RESULT(2) = norm( T*Z - Q'*B ) / (MAX(P,N)*norm(B)*ULP) *> RESULT(3) = norm( I - Q'*Q ) / ( M*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 complex_eig * * ===================================================================== SUBROUTINE CGQRTS( N, M, P, 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, P, N * .. * .. Array Arguments .. REAL RWORK( * ), RESULT( 4 ) COMPLEX A( LDA, * ), AF( LDA, * ), R( LDA, * ), $ Q( LDA, * ), B( LDB, * ), BF( LDB, * ), $ T( LDB, * ), Z( LDB, * ), BWK( LDB, * ), $ TAUA( * ), TAUB( * ), WORK( LWORK ) * .. * * ===================================================================== * * .. Parameters .. REAL ZERO, ONE PARAMETER ( ZERO = 0.0E+0, ONE = 1.0E+0 ) COMPLEX CZERO, CONE PARAMETER ( CZERO = ( 0.0E+0, 0.0E+0 ), $ CONE = ( 1.0E+0, 0.0E+0 ) ) COMPLEX CROGUE PARAMETER ( CROGUE = ( -1.0E+10, 0.0E+0 ) ) * .. * .. Local Scalars .. INTEGER INFO REAL ANORM, BNORM, ULP, UNFL, RESID * .. * .. External Functions .. REAL SLAMCH, CLANGE, CLANHE EXTERNAL SLAMCH, CLANGE, CLANHE * .. * .. External Subroutines .. EXTERNAL CGEMM, CLACPY, CLASET, CUNGQR, $ CUNGRQ, CHERK * .. * .. Intrinsic Functions .. INTRINSIC MAX, MIN, REAL * .. * .. Executable Statements .. * ULP = SLAMCH( 'Precision' ) UNFL = SLAMCH( 'Safe minimum' ) * * Copy the matrix A to the array AF. * CALL CLACPY( 'Full', N, M, A, LDA, AF, LDA ) CALL CLACPY( 'Full', N, P, B, LDB, BF, LDB ) * ANORM = MAX( CLANGE( '1', N, M, A, LDA, RWORK ), UNFL ) BNORM = MAX( CLANGE( '1', N, P, B, LDB, RWORK ), UNFL ) * * Factorize the matrices A and B in the arrays AF and BF. * CALL CGGQRF( N, M, P, AF, LDA, TAUA, BF, LDB, TAUB, WORK, $ LWORK, INFO ) * * Generate the N-by-N matrix Q * CALL CLASET( 'Full', N, N, CROGUE, CROGUE, Q, LDA ) CALL CLACPY( 'Lower', N-1, M, AF( 2,1 ), LDA, Q( 2,1 ), LDA ) CALL CUNGQR( N, N, MIN( N, M ), Q, LDA, TAUA, WORK, LWORK, INFO ) * * Generate the P-by-P matrix Z * CALL CLASET( 'Full', P, P, CROGUE, CROGUE, Z, LDB ) IF( N.LE.P ) THEN IF( N.GT.0 .AND. N.LT.P ) $ CALL CLACPY( 'Full', N, P-N, BF, LDB, Z( P-N+1, 1 ), LDB ) IF( N.GT.1 ) $ CALL CLACPY( 'Lower', N-1, N-1, BF( 2, P-N+1 ), LDB, $ Z( P-N+2, P-N+1 ), LDB ) ELSE IF( P.GT.1) $ CALL CLACPY( 'Lower', P-1, P-1, BF( N-P+2, 1 ), LDB, $ Z( 2, 1 ), LDB ) END IF CALL CUNGRQ( P, P, MIN( N, P ), Z, LDB, TAUB, WORK, LWORK, INFO ) * * Copy R * CALL CLASET( 'Full', N, M, CZERO, CZERO, R, LDA ) CALL CLACPY( 'Upper', N, M, AF, LDA, R, LDA ) * * Copy T * CALL CLASET( 'Full', N, P, CZERO, CZERO, T, LDB ) IF( N.LE.P ) THEN CALL CLACPY( 'Upper', N, N, BF( 1, P-N+1 ), LDB, T( 1, P-N+1 ), $ LDB ) ELSE CALL CLACPY( 'Full', N-P, P, BF, LDB, T, LDB ) CALL CLACPY( 'Upper', P, P, BF( N-P+1, 1 ), LDB, T( N-P+1, 1 ), $ LDB ) END IF * * Compute R - Q'*A * CALL CGEMM( 'Conjugate transpose', 'No transpose', N, M, N, -CONE, $ Q, LDA, A, LDA, CONE, R, LDA ) * * Compute norm( R - Q'*A ) / ( MAX(M,N)*norm(A)*ULP ) . * RESID = CLANGE( '1', N, M, R, LDA, RWORK ) IF( ANORM.GT.ZERO ) THEN RESULT( 1 ) = ( ( RESID / REAL( MAX(1,M,N) ) ) / ANORM ) / ULP ELSE RESULT( 1 ) = ZERO END IF * * Compute T*Z - Q'*B * CALL CGEMM( 'No Transpose', 'No transpose', N, P, P, CONE, T, LDB, $ Z, LDB, CZERO, BWK, LDB ) CALL CGEMM( 'Conjugate transpose', 'No transpose', N, P, N, -CONE, $ Q, LDA, B, LDB, CONE, BWK, LDB ) * * Compute norm( T*Z - Q'*B ) / ( MAX(P,N)*norm(A)*ULP ) . * RESID = CLANGE( '1', N, P, BWK, LDB, RWORK ) IF( BNORM.GT.ZERO ) THEN RESULT( 2 ) = ( ( RESID / REAL( MAX(1,P,N ) ) )/BNORM ) / ULP ELSE RESULT( 2 ) = ZERO END IF * * Compute I - Q'*Q * CALL CLASET( 'Full', N, N, CZERO, CONE, R, LDA ) CALL CHERK( 'Upper', 'Conjugate transpose', N, N, -ONE, Q, LDA, $ ONE, R, LDA ) * * Compute norm( I - Q'*Q ) / ( N * ULP ) . * RESID = CLANHE( '1', 'Upper', N, R, LDA, RWORK ) RESULT( 3 ) = ( RESID / REAL( MAX( 1, N ) ) ) / ULP * * Compute I - Z'*Z * CALL CLASET( 'Full', P, P, CZERO, CONE, T, LDB ) CALL CHERK( 'Upper', 'Conjugate transpose', P, P, -ONE, Z, LDB, $ ONE, T, LDB ) * * Compute norm( I - Z'*Z ) / ( P*ULP ) . * RESID = CLANHE( '1', 'Upper', P, T, LDB, RWORK ) RESULT( 4 ) = ( RESID / REAL( MAX( 1, P ) ) ) / ULP * RETURN * * End of CGQRTS * END