You can not select more than 25 topics Topics must start with a chinese character,a letter or number, can include dashes ('-') and can be up to 35 characters long.

dstt22.f 7.1 kB

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960616263646566676869707172737475767778798081828384858687888990919293949596979899100101102103104105106107108109110111112113114115116117118119120121122123124125126127128129130131132133134135136137138139140141142143144145146147148149150151152153154155156157158159160161162163164165166167168169170171172173174175176177178179180181182183184185186187188189190191192193194195196197198199200201202203204205206207208209210211212213214215216217218219220221222223224225226227228229230231232233234235236237238239240241242243244245246247248249250251252253254
  1. *> \brief \b DSTT22
  2. *
  3. * =========== DOCUMENTATION ===========
  4. *
  5. * Online html documentation available at
  6. * http://www.netlib.org/lapack/explore-html/
  7. *
  8. * Definition:
  9. * ===========
  10. *
  11. * SUBROUTINE DSTT22( N, M, KBAND, AD, AE, SD, SE, U, LDU, WORK,
  12. * LDWORK, RESULT )
  13. *
  14. * .. Scalar Arguments ..
  15. * INTEGER KBAND, LDU, LDWORK, M, N
  16. * ..
  17. * .. Array Arguments ..
  18. * DOUBLE PRECISION AD( * ), AE( * ), RESULT( 2 ), SD( * ),
  19. * $ SE( * ), U( LDU, * ), WORK( LDWORK, * )
  20. * ..
  21. *
  22. *
  23. *> \par Purpose:
  24. * =============
  25. *>
  26. *> \verbatim
  27. *>
  28. *> DSTT22 checks a set of M eigenvalues and eigenvectors,
  29. *>
  30. *> A U = U S
  31. *>
  32. *> where A is symmetric tridiagonal, the columns of U are orthogonal,
  33. *> and S is diagonal (if KBAND=0) or symmetric tridiagonal (if KBAND=1).
  34. *> Two tests are performed:
  35. *>
  36. *> RESULT(1) = | U' A U - S | / ( |A| m ulp )
  37. *>
  38. *> RESULT(2) = | I - U'U | / ( m ulp )
  39. *> \endverbatim
  40. *
  41. * Arguments:
  42. * ==========
  43. *
  44. *> \param[in] N
  45. *> \verbatim
  46. *> N is INTEGER
  47. *> The size of the matrix. If it is zero, DSTT22 does nothing.
  48. *> It must be at least zero.
  49. *> \endverbatim
  50. *>
  51. *> \param[in] M
  52. *> \verbatim
  53. *> M is INTEGER
  54. *> The number of eigenpairs to check. If it is zero, DSTT22
  55. *> does nothing. It must be at least zero.
  56. *> \endverbatim
  57. *>
  58. *> \param[in] KBAND
  59. *> \verbatim
  60. *> KBAND is INTEGER
  61. *> The bandwidth of the matrix S. It may only be zero or one.
  62. *> If zero, then S is diagonal, and SE is not referenced. If
  63. *> one, then S is symmetric tri-diagonal.
  64. *> \endverbatim
  65. *>
  66. *> \param[in] AD
  67. *> \verbatim
  68. *> AD is DOUBLE PRECISION array, dimension (N)
  69. *> The diagonal of the original (unfactored) matrix A. A is
  70. *> assumed to be symmetric tridiagonal.
  71. *> \endverbatim
  72. *>
  73. *> \param[in] AE
  74. *> \verbatim
  75. *> AE is DOUBLE PRECISION array, dimension (N)
  76. *> The off-diagonal of the original (unfactored) matrix A. A
  77. *> is assumed to be symmetric tridiagonal. AE(1) is ignored,
  78. *> AE(2) is the (1,2) and (2,1) element, etc.
  79. *> \endverbatim
  80. *>
  81. *> \param[in] SD
  82. *> \verbatim
  83. *> SD is DOUBLE PRECISION array, dimension (N)
  84. *> The diagonal of the (symmetric tri-) diagonal matrix S.
  85. *> \endverbatim
  86. *>
  87. *> \param[in] SE
  88. *> \verbatim
  89. *> SE is DOUBLE PRECISION array, dimension (N)
  90. *> The off-diagonal of the (symmetric tri-) diagonal matrix S.
  91. *> Not referenced if KBSND=0. If KBAND=1, then AE(1) is
  92. *> ignored, SE(2) is the (1,2) and (2,1) element, etc.
  93. *> \endverbatim
  94. *>
  95. *> \param[in] U
  96. *> \verbatim
  97. *> U is DOUBLE PRECISION array, dimension (LDU, N)
  98. *> The orthogonal matrix in the decomposition.
  99. *> \endverbatim
  100. *>
  101. *> \param[in] LDU
  102. *> \verbatim
  103. *> LDU is INTEGER
  104. *> The leading dimension of U. LDU must be at least N.
  105. *> \endverbatim
  106. *>
  107. *> \param[out] WORK
  108. *> \verbatim
  109. *> WORK is DOUBLE PRECISION array, dimension (LDWORK, M+1)
  110. *> \endverbatim
  111. *>
  112. *> \param[in] LDWORK
  113. *> \verbatim
  114. *> LDWORK is INTEGER
  115. *> The leading dimension of WORK. LDWORK must be at least
  116. *> max(1,M).
  117. *> \endverbatim
  118. *>
  119. *> \param[out] RESULT
  120. *> \verbatim
  121. *> RESULT is DOUBLE PRECISION array, dimension (2)
  122. *> The values computed by the two tests described above. The
  123. *> values are currently limited to 1/ulp, to avoid overflow.
  124. *> \endverbatim
  125. *
  126. * Authors:
  127. * ========
  128. *
  129. *> \author Univ. of Tennessee
  130. *> \author Univ. of California Berkeley
  131. *> \author Univ. of Colorado Denver
  132. *> \author NAG Ltd.
  133. *
  134. *> \date November 2011
  135. *
  136. *> \ingroup double_eig
  137. *
  138. * =====================================================================
  139. SUBROUTINE DSTT22( N, M, KBAND, AD, AE, SD, SE, U, LDU, WORK,
  140. $ LDWORK, RESULT )
  141. *
  142. * -- LAPACK test routine (version 3.4.0) --
  143. * -- LAPACK is a software package provided by Univ. of Tennessee, --
  144. * -- Univ. of California Berkeley, Univ. of Colorado Denver and NAG Ltd..--
  145. * November 2011
  146. *
  147. * .. Scalar Arguments ..
  148. INTEGER KBAND, LDU, LDWORK, M, N
  149. * ..
  150. * .. Array Arguments ..
  151. DOUBLE PRECISION AD( * ), AE( * ), RESULT( 2 ), SD( * ),
  152. $ SE( * ), U( LDU, * ), WORK( LDWORK, * )
  153. * ..
  154. *
  155. * =====================================================================
  156. *
  157. * .. Parameters ..
  158. DOUBLE PRECISION ZERO, ONE
  159. PARAMETER ( ZERO = 0.0D0, ONE = 1.0D0 )
  160. * ..
  161. * .. Local Scalars ..
  162. INTEGER I, J, K
  163. DOUBLE PRECISION ANORM, AUKJ, ULP, UNFL, WNORM
  164. * ..
  165. * .. External Functions ..
  166. DOUBLE PRECISION DLAMCH, DLANGE, DLANSY
  167. EXTERNAL DLAMCH, DLANGE, DLANSY
  168. * ..
  169. * .. External Subroutines ..
  170. EXTERNAL DGEMM
  171. * ..
  172. * .. Intrinsic Functions ..
  173. INTRINSIC ABS, DBLE, MAX, MIN
  174. * ..
  175. * .. Executable Statements ..
  176. *
  177. RESULT( 1 ) = ZERO
  178. RESULT( 2 ) = ZERO
  179. IF( N.LE.0 .OR. M.LE.0 )
  180. $ RETURN
  181. *
  182. UNFL = DLAMCH( 'Safe minimum' )
  183. ULP = DLAMCH( 'Epsilon' )
  184. *
  185. * Do Test 1
  186. *
  187. * Compute the 1-norm of A.
  188. *
  189. IF( N.GT.1 ) THEN
  190. ANORM = ABS( AD( 1 ) ) + ABS( AE( 1 ) )
  191. DO 10 J = 2, N - 1
  192. ANORM = MAX( ANORM, ABS( AD( J ) )+ABS( AE( J ) )+
  193. $ ABS( AE( J-1 ) ) )
  194. 10 CONTINUE
  195. ANORM = MAX( ANORM, ABS( AD( N ) )+ABS( AE( N-1 ) ) )
  196. ELSE
  197. ANORM = ABS( AD( 1 ) )
  198. END IF
  199. ANORM = MAX( ANORM, UNFL )
  200. *
  201. * Norm of U'AU - S
  202. *
  203. DO 40 I = 1, M
  204. DO 30 J = 1, M
  205. WORK( I, J ) = ZERO
  206. DO 20 K = 1, N
  207. AUKJ = AD( K )*U( K, J )
  208. IF( K.NE.N )
  209. $ AUKJ = AUKJ + AE( K )*U( K+1, J )
  210. IF( K.NE.1 )
  211. $ AUKJ = AUKJ + AE( K-1 )*U( K-1, J )
  212. WORK( I, J ) = WORK( I, J ) + U( K, I )*AUKJ
  213. 20 CONTINUE
  214. 30 CONTINUE
  215. WORK( I, I ) = WORK( I, I ) - SD( I )
  216. IF( KBAND.EQ.1 ) THEN
  217. IF( I.NE.1 )
  218. $ WORK( I, I-1 ) = WORK( I, I-1 ) - SE( I-1 )
  219. IF( I.NE.N )
  220. $ WORK( I, I+1 ) = WORK( I, I+1 ) - SE( I )
  221. END IF
  222. 40 CONTINUE
  223. *
  224. WNORM = DLANSY( '1', 'L', M, WORK, M, WORK( 1, M+1 ) )
  225. *
  226. IF( ANORM.GT.WNORM ) THEN
  227. RESULT( 1 ) = ( WNORM / ANORM ) / ( M*ULP )
  228. ELSE
  229. IF( ANORM.LT.ONE ) THEN
  230. RESULT( 1 ) = ( MIN( WNORM, M*ANORM ) / ANORM ) / ( M*ULP )
  231. ELSE
  232. RESULT( 1 ) = MIN( WNORM / ANORM, DBLE( M ) ) / ( M*ULP )
  233. END IF
  234. END IF
  235. *
  236. * Do Test 2
  237. *
  238. * Compute U'U - I
  239. *
  240. CALL DGEMM( 'T', 'N', M, M, N, ONE, U, LDU, U, LDU, ZERO, WORK,
  241. $ M )
  242. *
  243. DO 50 J = 1, M
  244. WORK( J, J ) = WORK( J, J ) - ONE
  245. 50 CONTINUE
  246. *
  247. RESULT( 2 ) = MIN( DBLE( M ), DLANGE( '1', M, M, WORK, M, WORK( 1,
  248. $ M+1 ) ) ) / ( M*ULP )
  249. *
  250. RETURN
  251. *
  252. * End of DSTT22
  253. *
  254. END