1: SUBROUTINE DSYTD2( UPLO, N, A, LDA, D, E, TAU, INFO )
2: *
3: * -- LAPACK routine (version 3.2) --
4: * -- LAPACK is a software package provided by Univ. of Tennessee, --
5: * -- Univ. of California Berkeley, Univ. of Colorado Denver and NAG Ltd..--
6: * November 2006
7: *
8: * .. Scalar Arguments ..
9: CHARACTER UPLO
10: INTEGER INFO, LDA, N
11: * ..
12: * .. Array Arguments ..
13: DOUBLE PRECISION A( LDA, * ), D( * ), E( * ), TAU( * )
14: * ..
15: *
16: * Purpose
17: * =======
18: *
19: * DSYTD2 reduces a real symmetric matrix A to symmetric tridiagonal
20: * form T by an orthogonal similarity transformation: Q' * A * Q = T.
21: *
22: * Arguments
23: * =========
24: *
25: * UPLO (input) CHARACTER*1
26: * Specifies whether the upper or lower triangular part of the
27: * symmetric matrix A is stored:
28: * = 'U': Upper triangular
29: * = 'L': Lower triangular
30: *
31: * N (input) INTEGER
32: * The order of the matrix A. N >= 0.
33: *
34: * A (input/output) DOUBLE PRECISION array, dimension (LDA,N)
35: * On entry, the symmetric matrix A. If UPLO = 'U', the leading
36: * n-by-n upper triangular part of A contains the upper
37: * triangular part of the matrix A, and the strictly lower
38: * triangular part of A is not referenced. If UPLO = 'L', the
39: * leading n-by-n lower triangular part of A contains the lower
40: * triangular part of the matrix A, and the strictly upper
41: * triangular part of A is not referenced.
42: * On exit, if UPLO = 'U', the diagonal and first superdiagonal
43: * of A are overwritten by the corresponding elements of the
44: * tridiagonal matrix T, and the elements above the first
45: * superdiagonal, with the array TAU, represent the orthogonal
46: * matrix Q as a product of elementary reflectors; if UPLO
47: * = 'L', the diagonal and first subdiagonal of A are over-
48: * written by the corresponding elements of the tridiagonal
49: * matrix T, and the elements below the first subdiagonal, with
50: * the array TAU, represent the orthogonal matrix Q as a product
51: * of elementary reflectors. See Further Details.
52: *
53: * LDA (input) INTEGER
54: * The leading dimension of the array A. LDA >= max(1,N).
55: *
56: * D (output) DOUBLE PRECISION array, dimension (N)
57: * The diagonal elements of the tridiagonal matrix T:
58: * D(i) = A(i,i).
59: *
60: * E (output) DOUBLE PRECISION array, dimension (N-1)
61: * The off-diagonal elements of the tridiagonal matrix T:
62: * E(i) = A(i,i+1) if UPLO = 'U', E(i) = A(i+1,i) if UPLO = 'L'.
63: *
64: * TAU (output) DOUBLE PRECISION array, dimension (N-1)
65: * The scalar factors of the elementary reflectors (see Further
66: * Details).
67: *
68: * INFO (output) INTEGER
69: * = 0: successful exit
70: * < 0: if INFO = -i, the i-th argument had an illegal value.
71: *
72: * Further Details
73: * ===============
74: *
75: * If UPLO = 'U', the matrix Q is represented as a product of elementary
76: * reflectors
77: *
78: * Q = H(n-1) . . . H(2) H(1).
79: *
80: * Each H(i) has the form
81: *
82: * H(i) = I - tau * v * v'
83: *
84: * where tau is a real scalar, and v is a real vector with
85: * v(i+1:n) = 0 and v(i) = 1; v(1:i-1) is stored on exit in
86: * A(1:i-1,i+1), and tau in TAU(i).
87: *
88: * If UPLO = 'L', the matrix Q is represented as a product of elementary
89: * reflectors
90: *
91: * Q = H(1) H(2) . . . H(n-1).
92: *
93: * Each H(i) has the form
94: *
95: * H(i) = I - tau * v * v'
96: *
97: * where tau is a real scalar, and v is a real vector with
98: * v(1:i) = 0 and v(i+1) = 1; v(i+2:n) is stored on exit in A(i+2:n,i),
99: * and tau in TAU(i).
100: *
101: * The contents of A on exit are illustrated by the following examples
102: * with n = 5:
103: *
104: * if UPLO = 'U': if UPLO = 'L':
105: *
106: * ( d e v2 v3 v4 ) ( d )
107: * ( d e v3 v4 ) ( e d )
108: * ( d e v4 ) ( v1 e d )
109: * ( d e ) ( v1 v2 e d )
110: * ( d ) ( v1 v2 v3 e d )
111: *
112: * where d and e denote diagonal and off-diagonal elements of T, and vi
113: * denotes an element of the vector defining H(i).
114: *
115: * =====================================================================
116: *
117: * .. Parameters ..
118: DOUBLE PRECISION ONE, ZERO, HALF
119: PARAMETER ( ONE = 1.0D0, ZERO = 0.0D0,
120: $ HALF = 1.0D0 / 2.0D0 )
121: * ..
122: * .. Local Scalars ..
123: LOGICAL UPPER
124: INTEGER I
125: DOUBLE PRECISION ALPHA, TAUI
126: * ..
127: * .. External Subroutines ..
128: EXTERNAL DAXPY, DLARFG, DSYMV, DSYR2, XERBLA
129: * ..
130: * .. External Functions ..
131: LOGICAL LSAME
132: DOUBLE PRECISION DDOT
133: EXTERNAL LSAME, DDOT
134: * ..
135: * .. Intrinsic Functions ..
136: INTRINSIC MAX, MIN
137: * ..
138: * .. Executable Statements ..
139: *
140: * Test the input parameters
141: *
142: INFO = 0
143: UPPER = LSAME( UPLO, 'U' )
144: IF( .NOT.UPPER .AND. .NOT.LSAME( UPLO, 'L' ) ) THEN
145: INFO = -1
146: ELSE IF( N.LT.0 ) THEN
147: INFO = -2
148: ELSE IF( LDA.LT.MAX( 1, N ) ) THEN
149: INFO = -4
150: END IF
151: IF( INFO.NE.0 ) THEN
152: CALL XERBLA( 'DSYTD2', -INFO )
153: RETURN
154: END IF
155: *
156: * Quick return if possible
157: *
158: IF( N.LE.0 )
159: $ RETURN
160: *
161: IF( UPPER ) THEN
162: *
163: * Reduce the upper triangle of A
164: *
165: DO 10 I = N - 1, 1, -1
166: *
167: * Generate elementary reflector H(i) = I - tau * v * v'
168: * to annihilate A(1:i-1,i+1)
169: *
170: CALL DLARFG( I, A( I, I+1 ), A( 1, I+1 ), 1, TAUI )
171: E( I ) = A( I, I+1 )
172: *
173: IF( TAUI.NE.ZERO ) THEN
174: *
175: * Apply H(i) from both sides to A(1:i,1:i)
176: *
177: A( I, I+1 ) = ONE
178: *
179: * Compute x := tau * A * v storing x in TAU(1:i)
180: *
181: CALL DSYMV( UPLO, I, TAUI, A, LDA, A( 1, I+1 ), 1, ZERO,
182: $ TAU, 1 )
183: *
184: * Compute w := x - 1/2 * tau * (x'*v) * v
185: *
186: ALPHA = -HALF*TAUI*DDOT( I, TAU, 1, A( 1, I+1 ), 1 )
187: CALL DAXPY( I, ALPHA, A( 1, I+1 ), 1, TAU, 1 )
188: *
189: * Apply the transformation as a rank-2 update:
190: * A := A - v * w' - w * v'
191: *
192: CALL DSYR2( UPLO, I, -ONE, A( 1, I+1 ), 1, TAU, 1, A,
193: $ LDA )
194: *
195: A( I, I+1 ) = E( I )
196: END IF
197: D( I+1 ) = A( I+1, I+1 )
198: TAU( I ) = TAUI
199: 10 CONTINUE
200: D( 1 ) = A( 1, 1 )
201: ELSE
202: *
203: * Reduce the lower triangle of A
204: *
205: DO 20 I = 1, N - 1
206: *
207: * Generate elementary reflector H(i) = I - tau * v * v'
208: * to annihilate A(i+2:n,i)
209: *
210: CALL DLARFG( N-I, A( I+1, I ), A( MIN( I+2, N ), I ), 1,
211: $ TAUI )
212: E( I ) = A( I+1, I )
213: *
214: IF( TAUI.NE.ZERO ) THEN
215: *
216: * Apply H(i) from both sides to A(i+1:n,i+1:n)
217: *
218: A( I+1, I ) = ONE
219: *
220: * Compute x := tau * A * v storing y in TAU(i:n-1)
221: *
222: CALL DSYMV( UPLO, N-I, TAUI, A( I+1, I+1 ), LDA,
223: $ A( I+1, I ), 1, ZERO, TAU( I ), 1 )
224: *
225: * Compute w := x - 1/2 * tau * (x'*v) * v
226: *
227: ALPHA = -HALF*TAUI*DDOT( N-I, TAU( I ), 1, A( I+1, I ),
228: $ 1 )
229: CALL DAXPY( N-I, ALPHA, A( I+1, I ), 1, TAU( I ), 1 )
230: *
231: * Apply the transformation as a rank-2 update:
232: * A := A - v * w' - w * v'
233: *
234: CALL DSYR2( UPLO, N-I, -ONE, A( I+1, I ), 1, TAU( I ), 1,
235: $ A( I+1, I+1 ), LDA )
236: *
237: A( I+1, I ) = E( I )
238: END IF
239: D( I ) = A( I, I )
240: TAU( I ) = TAUI
241: 20 CONTINUE
242: D( N ) = A( N, N )
243: END IF
244: *
245: RETURN
246: *
247: * End of DSYTD2
248: *
249: END
CVSweb interface <joel.bertrand@systella.fr>