LAPACK 3.3.0
|
00001 SUBROUTINE ZPOT01( UPLO, N, A, LDA, AFAC, LDAFAC, RWORK, RESID ) 00002 * 00003 * -- LAPACK test routine (version 3.1) -- 00004 * Univ. of Tennessee, Univ. of California Berkeley and NAG Ltd.. 00005 * November 2006 00006 * 00007 * .. Scalar Arguments .. 00008 CHARACTER UPLO 00009 INTEGER LDA, LDAFAC, N 00010 DOUBLE PRECISION RESID 00011 * .. 00012 * .. Array Arguments .. 00013 DOUBLE PRECISION RWORK( * ) 00014 COMPLEX*16 A( LDA, * ), AFAC( LDAFAC, * ) 00015 * .. 00016 * 00017 * Purpose 00018 * ======= 00019 * 00020 * ZPOT01 reconstructs a Hermitian positive definite matrix A from 00021 * its L*L' or U'*U factorization and computes the residual 00022 * norm( L*L' - A ) / ( N * norm(A) * EPS ) or 00023 * norm( U'*U - A ) / ( N * norm(A) * EPS ), 00024 * where EPS is the machine epsilon, L' is the conjugate transpose of L, 00025 * and U' is the conjugate transpose of U. 00026 * 00027 * Arguments 00028 * ========== 00029 * 00030 * UPLO (input) CHARACTER*1 00031 * Specifies whether the upper or lower triangular part of the 00032 * Hermitian matrix A is stored: 00033 * = 'U': Upper triangular 00034 * = 'L': Lower triangular 00035 * 00036 * N (input) INTEGER 00037 * The number of rows and columns of the matrix A. N >= 0. 00038 * 00039 * A (input) COMPLEX*16 array, dimension (LDA,N) 00040 * The original Hermitian matrix A. 00041 * 00042 * LDA (input) INTEGER 00043 * The leading dimension of the array A. LDA >= max(1,N) 00044 * 00045 * AFAC (input/output) COMPLEX*16 array, dimension (LDAFAC,N) 00046 * On entry, the factor L or U from the L*L' or U'*U 00047 * factorization of A. 00048 * Overwritten with the reconstructed matrix, and then with the 00049 * difference L*L' - A (or U'*U - A). 00050 * 00051 * LDAFAC (input) INTEGER 00052 * The leading dimension of the array AFAC. LDAFAC >= max(1,N). 00053 * 00054 * RWORK (workspace) DOUBLE PRECISION array, dimension (N) 00055 * 00056 * RESID (output) DOUBLE PRECISION 00057 * If UPLO = 'L', norm(L*L' - A) / ( N * norm(A) * EPS ) 00058 * If UPLO = 'U', norm(U'*U - A) / ( N * norm(A) * EPS ) 00059 * 00060 * ===================================================================== 00061 * 00062 * .. Parameters .. 00063 DOUBLE PRECISION ZERO, ONE 00064 PARAMETER ( ZERO = 0.0D+0, ONE = 1.0D+0 ) 00065 * .. 00066 * .. Local Scalars .. 00067 INTEGER I, J, K 00068 DOUBLE PRECISION ANORM, EPS, TR 00069 COMPLEX*16 TC 00070 * .. 00071 * .. External Functions .. 00072 LOGICAL LSAME 00073 DOUBLE PRECISION DLAMCH, ZLANHE 00074 COMPLEX*16 ZDOTC 00075 EXTERNAL LSAME, DLAMCH, ZLANHE, ZDOTC 00076 * .. 00077 * .. External Subroutines .. 00078 EXTERNAL ZHER, ZSCAL, ZTRMV 00079 * .. 00080 * .. Intrinsic Functions .. 00081 INTRINSIC DBLE, DIMAG 00082 * .. 00083 * .. Executable Statements .. 00084 * 00085 * Quick exit if N = 0. 00086 * 00087 IF( N.LE.0 ) THEN 00088 RESID = ZERO 00089 RETURN 00090 END IF 00091 * 00092 * Exit with RESID = 1/EPS if ANORM = 0. 00093 * 00094 EPS = DLAMCH( 'Epsilon' ) 00095 ANORM = ZLANHE( '1', UPLO, N, A, LDA, RWORK ) 00096 IF( ANORM.LE.ZERO ) THEN 00097 RESID = ONE / EPS 00098 RETURN 00099 END IF 00100 * 00101 * Check the imaginary parts of the diagonal elements and return with 00102 * an error code if any are nonzero. 00103 * 00104 DO 10 J = 1, N 00105 IF( DIMAG( AFAC( J, J ) ).NE.ZERO ) THEN 00106 RESID = ONE / EPS 00107 RETURN 00108 END IF 00109 10 CONTINUE 00110 * 00111 * Compute the product U'*U, overwriting U. 00112 * 00113 IF( LSAME( UPLO, 'U' ) ) THEN 00114 DO 20 K = N, 1, -1 00115 * 00116 * Compute the (K,K) element of the result. 00117 * 00118 TR = ZDOTC( K, AFAC( 1, K ), 1, AFAC( 1, K ), 1 ) 00119 AFAC( K, K ) = TR 00120 * 00121 * Compute the rest of column K. 00122 * 00123 CALL ZTRMV( 'Upper', 'Conjugate', 'Non-unit', K-1, AFAC, 00124 $ LDAFAC, AFAC( 1, K ), 1 ) 00125 * 00126 20 CONTINUE 00127 * 00128 * Compute the product L*L', overwriting L. 00129 * 00130 ELSE 00131 DO 30 K = N, 1, -1 00132 * 00133 * Add a multiple of column K of the factor L to each of 00134 * columns K+1 through N. 00135 * 00136 IF( K+1.LE.N ) 00137 $ CALL ZHER( 'Lower', N-K, ONE, AFAC( K+1, K ), 1, 00138 $ AFAC( K+1, K+1 ), LDAFAC ) 00139 * 00140 * Scale column K by the diagonal element. 00141 * 00142 TC = AFAC( K, K ) 00143 CALL ZSCAL( N-K+1, TC, AFAC( K, K ), 1 ) 00144 * 00145 30 CONTINUE 00146 END IF 00147 * 00148 * Compute the difference L*L' - A (or U'*U - A). 00149 * 00150 IF( LSAME( UPLO, 'U' ) ) THEN 00151 DO 50 J = 1, N 00152 DO 40 I = 1, J - 1 00153 AFAC( I, J ) = AFAC( I, J ) - A( I, J ) 00154 40 CONTINUE 00155 AFAC( J, J ) = AFAC( J, J ) - DBLE( A( J, J ) ) 00156 50 CONTINUE 00157 ELSE 00158 DO 70 J = 1, N 00159 AFAC( J, J ) = AFAC( J, J ) - DBLE( A( J, J ) ) 00160 DO 60 I = J + 1, N 00161 AFAC( I, J ) = AFAC( I, J ) - A( I, J ) 00162 60 CONTINUE 00163 70 CONTINUE 00164 END IF 00165 * 00166 * Compute norm( L*U - A ) / ( N * norm(A) * EPS ) 00167 * 00168 RESID = ZLANHE( '1', UPLO, N, AFAC, LDAFAC, RWORK ) 00169 * 00170 RESID = ( ( RESID / DBLE( N ) ) / ANORM ) / EPS 00171 * 00172 RETURN 00173 * 00174 * End of ZPOT01 00175 * 00176 END