LAPACK 3.3.0
|
00001 SUBROUTINE DPPTRF( UPLO, N, AP, INFO ) 00002 * 00003 * -- LAPACK routine (version 3.2) -- 00004 * -- LAPACK is a software package provided by Univ. of Tennessee, -- 00005 * -- Univ. of California Berkeley, Univ. of Colorado Denver and NAG Ltd..-- 00006 * November 2006 00007 * 00008 * .. Scalar Arguments .. 00009 CHARACTER UPLO 00010 INTEGER INFO, N 00011 * .. 00012 * .. Array Arguments .. 00013 DOUBLE PRECISION AP( * ) 00014 * .. 00015 * 00016 * Purpose 00017 * ======= 00018 * 00019 * DPPTRF computes the Cholesky factorization of a real symmetric 00020 * positive definite matrix A stored in packed format. 00021 * 00022 * The factorization has the form 00023 * A = U**T * U, if UPLO = 'U', or 00024 * A = L * L**T, if UPLO = 'L', 00025 * where U is an upper triangular matrix and L is lower triangular. 00026 * 00027 * Arguments 00028 * ========= 00029 * 00030 * UPLO (input) CHARACTER*1 00031 * = 'U': Upper triangle of A is stored; 00032 * = 'L': Lower triangle of A is stored. 00033 * 00034 * N (input) INTEGER 00035 * The order of the matrix A. N >= 0. 00036 * 00037 * AP (input/output) DOUBLE PRECISION array, dimension (N*(N+1)/2) 00038 * On entry, the upper or lower triangle of the symmetric matrix 00039 * A, packed columnwise in a linear array. The j-th column of A 00040 * is stored in the array AP as follows: 00041 * if UPLO = 'U', AP(i + (j-1)*j/2) = A(i,j) for 1<=i<=j; 00042 * if UPLO = 'L', AP(i + (j-1)*(2n-j)/2) = A(i,j) for j<=i<=n. 00043 * See below for further details. 00044 * 00045 * On exit, if INFO = 0, the triangular factor U or L from the 00046 * Cholesky factorization A = U**T*U or A = L*L**T, in the same 00047 * storage format as A. 00048 * 00049 * INFO (output) INTEGER 00050 * = 0: successful exit 00051 * < 0: if INFO = -i, the i-th argument had an illegal value 00052 * > 0: if INFO = i, the leading minor of order i is not 00053 * positive definite, and the factorization could not be 00054 * completed. 00055 * 00056 * Further Details 00057 * ======= ======= 00058 * 00059 * The packed storage scheme is illustrated by the following example 00060 * when N = 4, UPLO = 'U': 00061 * 00062 * Two-dimensional storage of the symmetric matrix A: 00063 * 00064 * a11 a12 a13 a14 00065 * a22 a23 a24 00066 * a33 a34 (aij = aji) 00067 * a44 00068 * 00069 * Packed storage of the upper triangle of A: 00070 * 00071 * AP = [ a11, a12, a22, a13, a23, a33, a14, a24, a34, a44 ] 00072 * 00073 * ===================================================================== 00074 * 00075 * .. Parameters .. 00076 DOUBLE PRECISION ONE, ZERO 00077 PARAMETER ( ONE = 1.0D+0, ZERO = 0.0D+0 ) 00078 * .. 00079 * .. Local Scalars .. 00080 LOGICAL UPPER 00081 INTEGER J, JC, JJ 00082 DOUBLE PRECISION AJJ 00083 * .. 00084 * .. External Functions .. 00085 LOGICAL LSAME 00086 DOUBLE PRECISION DDOT 00087 EXTERNAL LSAME, DDOT 00088 * .. 00089 * .. External Subroutines .. 00090 EXTERNAL DSCAL, DSPR, DTPSV, XERBLA 00091 * .. 00092 * .. Intrinsic Functions .. 00093 INTRINSIC SQRT 00094 * .. 00095 * .. Executable Statements .. 00096 * 00097 * Test the input parameters. 00098 * 00099 INFO = 0 00100 UPPER = LSAME( UPLO, 'U' ) 00101 IF( .NOT.UPPER .AND. .NOT.LSAME( UPLO, 'L' ) ) THEN 00102 INFO = -1 00103 ELSE IF( N.LT.0 ) THEN 00104 INFO = -2 00105 END IF 00106 IF( INFO.NE.0 ) THEN 00107 CALL XERBLA( 'DPPTRF', -INFO ) 00108 RETURN 00109 END IF 00110 * 00111 * Quick return if possible 00112 * 00113 IF( N.EQ.0 ) 00114 $ RETURN 00115 * 00116 IF( UPPER ) THEN 00117 * 00118 * Compute the Cholesky factorization A = U'*U. 00119 * 00120 JJ = 0 00121 DO 10 J = 1, N 00122 JC = JJ + 1 00123 JJ = JJ + J 00124 * 00125 * Compute elements 1:J-1 of column J. 00126 * 00127 IF( J.GT.1 ) 00128 $ CALL DTPSV( 'Upper', 'Transpose', 'Non-unit', J-1, AP, 00129 $ AP( JC ), 1 ) 00130 * 00131 * Compute U(J,J) and test for non-positive-definiteness. 00132 * 00133 AJJ = AP( JJ ) - DDOT( J-1, AP( JC ), 1, AP( JC ), 1 ) 00134 IF( AJJ.LE.ZERO ) THEN 00135 AP( JJ ) = AJJ 00136 GO TO 30 00137 END IF 00138 AP( JJ ) = SQRT( AJJ ) 00139 10 CONTINUE 00140 ELSE 00141 * 00142 * Compute the Cholesky factorization A = L*L'. 00143 * 00144 JJ = 1 00145 DO 20 J = 1, N 00146 * 00147 * Compute L(J,J) and test for non-positive-definiteness. 00148 * 00149 AJJ = AP( JJ ) 00150 IF( AJJ.LE.ZERO ) THEN 00151 AP( JJ ) = AJJ 00152 GO TO 30 00153 END IF 00154 AJJ = SQRT( AJJ ) 00155 AP( JJ ) = AJJ 00156 * 00157 * Compute elements J+1:N of column J and update the trailing 00158 * submatrix. 00159 * 00160 IF( J.LT.N ) THEN 00161 CALL DSCAL( N-J, ONE / AJJ, AP( JJ+1 ), 1 ) 00162 CALL DSPR( 'Lower', N-J, -ONE, AP( JJ+1 ), 1, 00163 $ AP( JJ+N-J+1 ) ) 00164 JJ = JJ + N - J + 1 00165 END IF 00166 20 CONTINUE 00167 END IF 00168 GO TO 40 00169 * 00170 30 CONTINUE 00171 INFO = J 00172 * 00173 40 CONTINUE 00174 RETURN 00175 * 00176 * End of DPPTRF 00177 * 00178 END