LAPACK  3.10.0
LAPACK: Linear Algebra PACKage

◆ chetrs()

subroutine chetrs ( character  UPLO,
integer  N,
integer  NRHS,
complex, dimension( lda, * )  A,
integer  LDA,
integer, dimension( * )  IPIV,
complex, dimension( ldb, * )  B,
integer  LDB,
integer  INFO 
)

CHETRS

Download CHETRS + dependencies [TGZ] [ZIP] [TXT]

Purpose:
 CHETRS solves a system of linear equations A*X = B with a complex
 Hermitian matrix A using the factorization A = U*D*U**H or
 A = L*D*L**H computed by CHETRF.
Parameters
[in]UPLO
          UPLO is CHARACTER*1
          Specifies whether the details of the factorization are stored
          as an upper or lower triangular matrix.
          = 'U':  Upper triangular, form is A = U*D*U**H;
          = 'L':  Lower triangular, form is A = L*D*L**H.
[in]N
          N is INTEGER
          The order of the matrix A.  N >= 0.
[in]NRHS
          NRHS is INTEGER
          The number of right hand sides, i.e., the number of columns
          of the matrix B.  NRHS >= 0.
[in]A
          A is COMPLEX array, dimension (LDA,N)
          The block diagonal matrix D and the multipliers used to
          obtain the factor U or L as computed by CHETRF.
[in]LDA
          LDA is INTEGER
          The leading dimension of the array A.  LDA >= max(1,N).
[in]IPIV
          IPIV is INTEGER array, dimension (N)
          Details of the interchanges and the block structure of D
          as determined by CHETRF.
[in,out]B
          B is COMPLEX array, dimension (LDB,NRHS)
          On entry, the right hand side matrix B.
          On exit, the solution matrix X.
[in]LDB
          LDB is INTEGER
          The leading dimension of the array B.  LDB >= max(1,N).
[out]INFO
          INFO is INTEGER
          = 0:  successful exit
          < 0:  if INFO = -i, the i-th argument had an illegal value
Author
Univ. of Tennessee
Univ. of California Berkeley
Univ. of Colorado Denver
NAG Ltd.

Definition at line 119 of file chetrs.f.

120 *
121 * -- LAPACK computational routine --
122 * -- LAPACK is a software package provided by Univ. of Tennessee, --
123 * -- Univ. of California Berkeley, Univ. of Colorado Denver and NAG Ltd..--
124 *
125 * .. Scalar Arguments ..
126  CHARACTER UPLO
127  INTEGER INFO, LDA, LDB, N, NRHS
128 * ..
129 * .. Array Arguments ..
130  INTEGER IPIV( * )
131  COMPLEX A( LDA, * ), B( LDB, * )
132 * ..
133 *
134 * =====================================================================
135 *
136 * .. Parameters ..
137  COMPLEX ONE
138  parameter( one = ( 1.0e+0, 0.0e+0 ) )
139 * ..
140 * .. Local Scalars ..
141  LOGICAL UPPER
142  INTEGER J, K, KP
143  REAL S
144  COMPLEX AK, AKM1, AKM1K, BK, BKM1, DENOM
145 * ..
146 * .. External Functions ..
147  LOGICAL LSAME
148  EXTERNAL lsame
149 * ..
150 * .. External Subroutines ..
151  EXTERNAL cgemv, cgeru, clacgv, csscal, cswap, xerbla
152 * ..
153 * .. Intrinsic Functions ..
154  INTRINSIC conjg, max, real
155 * ..
156 * .. Executable Statements ..
157 *
158  info = 0
159  upper = lsame( uplo, 'U' )
160  IF( .NOT.upper .AND. .NOT.lsame( uplo, 'L' ) ) THEN
161  info = -1
162  ELSE IF( n.LT.0 ) THEN
163  info = -2
164  ELSE IF( nrhs.LT.0 ) THEN
165  info = -3
166  ELSE IF( lda.LT.max( 1, n ) ) THEN
167  info = -5
168  ELSE IF( ldb.LT.max( 1, n ) ) THEN
169  info = -8
170  END IF
171  IF( info.NE.0 ) THEN
172  CALL xerbla( 'CHETRS', -info )
173  RETURN
174  END IF
175 *
176 * Quick return if possible
177 *
178  IF( n.EQ.0 .OR. nrhs.EQ.0 )
179  $ RETURN
180 *
181  IF( upper ) THEN
182 *
183 * Solve A*X = B, where A = U*D*U**H.
184 *
185 * First solve U*D*X = B, overwriting B with X.
186 *
187 * K is the main loop index, decreasing from N to 1 in steps of
188 * 1 or 2, depending on the size of the diagonal blocks.
189 *
190  k = n
191  10 CONTINUE
192 *
193 * If K < 1, exit from loop.
194 *
195  IF( k.LT.1 )
196  $ GO TO 30
197 *
198  IF( ipiv( k ).GT.0 ) THEN
199 *
200 * 1 x 1 diagonal block
201 *
202 * Interchange rows K and IPIV(K).
203 *
204  kp = ipiv( k )
205  IF( kp.NE.k )
206  $ CALL cswap( nrhs, b( k, 1 ), ldb, b( kp, 1 ), ldb )
207 *
208 * Multiply by inv(U(K)), where U(K) is the transformation
209 * stored in column K of A.
210 *
211  CALL cgeru( k-1, nrhs, -one, a( 1, k ), 1, b( k, 1 ), ldb,
212  $ b( 1, 1 ), ldb )
213 *
214 * Multiply by the inverse of the diagonal block.
215 *
216  s = real( one ) / real( a( k, k ) )
217  CALL csscal( nrhs, s, b( k, 1 ), ldb )
218  k = k - 1
219  ELSE
220 *
221 * 2 x 2 diagonal block
222 *
223 * Interchange rows K-1 and -IPIV(K).
224 *
225  kp = -ipiv( k )
226  IF( kp.NE.k-1 )
227  $ CALL cswap( nrhs, b( k-1, 1 ), ldb, b( kp, 1 ), ldb )
228 *
229 * Multiply by inv(U(K)), where U(K) is the transformation
230 * stored in columns K-1 and K of A.
231 *
232  CALL cgeru( k-2, nrhs, -one, a( 1, k ), 1, b( k, 1 ), ldb,
233  $ b( 1, 1 ), ldb )
234  CALL cgeru( k-2, nrhs, -one, a( 1, k-1 ), 1, b( k-1, 1 ),
235  $ ldb, b( 1, 1 ), ldb )
236 *
237 * Multiply by the inverse of the diagonal block.
238 *
239  akm1k = a( k-1, k )
240  akm1 = a( k-1, k-1 ) / akm1k
241  ak = a( k, k ) / conjg( akm1k )
242  denom = akm1*ak - one
243  DO 20 j = 1, nrhs
244  bkm1 = b( k-1, j ) / akm1k
245  bk = b( k, j ) / conjg( akm1k )
246  b( k-1, j ) = ( ak*bkm1-bk ) / denom
247  b( k, j ) = ( akm1*bk-bkm1 ) / denom
248  20 CONTINUE
249  k = k - 2
250  END IF
251 *
252  GO TO 10
253  30 CONTINUE
254 *
255 * Next solve U**H *X = B, overwriting B with X.
256 *
257 * K is the main loop index, increasing from 1 to N in steps of
258 * 1 or 2, depending on the size of the diagonal blocks.
259 *
260  k = 1
261  40 CONTINUE
262 *
263 * If K > N, exit from loop.
264 *
265  IF( k.GT.n )
266  $ GO TO 50
267 *
268  IF( ipiv( k ).GT.0 ) THEN
269 *
270 * 1 x 1 diagonal block
271 *
272 * Multiply by inv(U**H(K)), where U(K) is the transformation
273 * stored in column K of A.
274 *
275  IF( k.GT.1 ) THEN
276  CALL clacgv( nrhs, b( k, 1 ), ldb )
277  CALL cgemv( 'Conjugate transpose', k-1, nrhs, -one, b,
278  $ ldb, a( 1, k ), 1, one, b( k, 1 ), ldb )
279  CALL clacgv( nrhs, b( k, 1 ), ldb )
280  END IF
281 *
282 * Interchange rows K and IPIV(K).
283 *
284  kp = ipiv( k )
285  IF( kp.NE.k )
286  $ CALL cswap( nrhs, b( k, 1 ), ldb, b( kp, 1 ), ldb )
287  k = k + 1
288  ELSE
289 *
290 * 2 x 2 diagonal block
291 *
292 * Multiply by inv(U**H(K+1)), where U(K+1) is the transformation
293 * stored in columns K and K+1 of A.
294 *
295  IF( k.GT.1 ) THEN
296  CALL clacgv( nrhs, b( k, 1 ), ldb )
297  CALL cgemv( 'Conjugate transpose', k-1, nrhs, -one, b,
298  $ ldb, a( 1, k ), 1, one, b( k, 1 ), ldb )
299  CALL clacgv( nrhs, b( k, 1 ), ldb )
300 *
301  CALL clacgv( nrhs, b( k+1, 1 ), ldb )
302  CALL cgemv( 'Conjugate transpose', k-1, nrhs, -one, b,
303  $ ldb, a( 1, k+1 ), 1, one, b( k+1, 1 ), ldb )
304  CALL clacgv( nrhs, b( k+1, 1 ), ldb )
305  END IF
306 *
307 * Interchange rows K and -IPIV(K).
308 *
309  kp = -ipiv( k )
310  IF( kp.NE.k )
311  $ CALL cswap( nrhs, b( k, 1 ), ldb, b( kp, 1 ), ldb )
312  k = k + 2
313  END IF
314 *
315  GO TO 40
316  50 CONTINUE
317 *
318  ELSE
319 *
320 * Solve A*X = B, where A = L*D*L**H.
321 *
322 * First solve L*D*X = B, overwriting B with X.
323 *
324 * K is the main loop index, increasing from 1 to N in steps of
325 * 1 or 2, depending on the size of the diagonal blocks.
326 *
327  k = 1
328  60 CONTINUE
329 *
330 * If K > N, exit from loop.
331 *
332  IF( k.GT.n )
333  $ GO TO 80
334 *
335  IF( ipiv( k ).GT.0 ) THEN
336 *
337 * 1 x 1 diagonal block
338 *
339 * Interchange rows K and IPIV(K).
340 *
341  kp = ipiv( k )
342  IF( kp.NE.k )
343  $ CALL cswap( nrhs, b( k, 1 ), ldb, b( kp, 1 ), ldb )
344 *
345 * Multiply by inv(L(K)), where L(K) is the transformation
346 * stored in column K of A.
347 *
348  IF( k.LT.n )
349  $ CALL cgeru( n-k, nrhs, -one, a( k+1, k ), 1, b( k, 1 ),
350  $ ldb, b( k+1, 1 ), ldb )
351 *
352 * Multiply by the inverse of the diagonal block.
353 *
354  s = real( one ) / real( a( k, k ) )
355  CALL csscal( nrhs, s, b( k, 1 ), ldb )
356  k = k + 1
357  ELSE
358 *
359 * 2 x 2 diagonal block
360 *
361 * Interchange rows K+1 and -IPIV(K).
362 *
363  kp = -ipiv( k )
364  IF( kp.NE.k+1 )
365  $ CALL cswap( nrhs, b( k+1, 1 ), ldb, b( kp, 1 ), ldb )
366 *
367 * Multiply by inv(L(K)), where L(K) is the transformation
368 * stored in columns K and K+1 of A.
369 *
370  IF( k.LT.n-1 ) THEN
371  CALL cgeru( n-k-1, nrhs, -one, a( k+2, k ), 1, b( k, 1 ),
372  $ ldb, b( k+2, 1 ), ldb )
373  CALL cgeru( n-k-1, nrhs, -one, a( k+2, k+1 ), 1,
374  $ b( k+1, 1 ), ldb, b( k+2, 1 ), ldb )
375  END IF
376 *
377 * Multiply by the inverse of the diagonal block.
378 *
379  akm1k = a( k+1, k )
380  akm1 = a( k, k ) / conjg( akm1k )
381  ak = a( k+1, k+1 ) / akm1k
382  denom = akm1*ak - one
383  DO 70 j = 1, nrhs
384  bkm1 = b( k, j ) / conjg( akm1k )
385  bk = b( k+1, j ) / akm1k
386  b( k, j ) = ( ak*bkm1-bk ) / denom
387  b( k+1, j ) = ( akm1*bk-bkm1 ) / denom
388  70 CONTINUE
389  k = k + 2
390  END IF
391 *
392  GO TO 60
393  80 CONTINUE
394 *
395 * Next solve L**H *X = B, overwriting B with X.
396 *
397 * K is the main loop index, decreasing from N to 1 in steps of
398 * 1 or 2, depending on the size of the diagonal blocks.
399 *
400  k = n
401  90 CONTINUE
402 *
403 * If K < 1, exit from loop.
404 *
405  IF( k.LT.1 )
406  $ GO TO 100
407 *
408  IF( ipiv( k ).GT.0 ) THEN
409 *
410 * 1 x 1 diagonal block
411 *
412 * Multiply by inv(L**H(K)), where L(K) is the transformation
413 * stored in column K of A.
414 *
415  IF( k.LT.n ) THEN
416  CALL clacgv( nrhs, b( k, 1 ), ldb )
417  CALL cgemv( 'Conjugate transpose', n-k, nrhs, -one,
418  $ b( k+1, 1 ), ldb, a( k+1, k ), 1, one,
419  $ b( k, 1 ), ldb )
420  CALL clacgv( nrhs, b( k, 1 ), ldb )
421  END IF
422 *
423 * Interchange rows K and IPIV(K).
424 *
425  kp = ipiv( k )
426  IF( kp.NE.k )
427  $ CALL cswap( nrhs, b( k, 1 ), ldb, b( kp, 1 ), ldb )
428  k = k - 1
429  ELSE
430 *
431 * 2 x 2 diagonal block
432 *
433 * Multiply by inv(L**H(K-1)), where L(K-1) is the transformation
434 * stored in columns K-1 and K of A.
435 *
436  IF( k.LT.n ) THEN
437  CALL clacgv( nrhs, b( k, 1 ), ldb )
438  CALL cgemv( 'Conjugate transpose', n-k, nrhs, -one,
439  $ b( k+1, 1 ), ldb, a( k+1, k ), 1, one,
440  $ b( k, 1 ), ldb )
441  CALL clacgv( nrhs, b( k, 1 ), ldb )
442 *
443  CALL clacgv( nrhs, b( k-1, 1 ), ldb )
444  CALL cgemv( 'Conjugate transpose', n-k, nrhs, -one,
445  $ b( k+1, 1 ), ldb, a( k+1, k-1 ), 1, one,
446  $ b( k-1, 1 ), ldb )
447  CALL clacgv( nrhs, b( k-1, 1 ), ldb )
448  END IF
449 *
450 * Interchange rows K and -IPIV(K).
451 *
452  kp = -ipiv( k )
453  IF( kp.NE.k )
454  $ CALL cswap( nrhs, b( k, 1 ), ldb, b( kp, 1 ), ldb )
455  k = k - 2
456  END IF
457 *
458  GO TO 90
459  100 CONTINUE
460  END IF
461 *
462  RETURN
463 *
464 * End of CHETRS
465 *
subroutine xerbla(SRNAME, INFO)
XERBLA
Definition: xerbla.f:60
logical function lsame(CA, CB)
LSAME
Definition: lsame.f:53
subroutine csscal(N, SA, CX, INCX)
CSSCAL
Definition: csscal.f:78
subroutine cswap(N, CX, INCX, CY, INCY)
CSWAP
Definition: cswap.f:81
subroutine cgemv(TRANS, M, N, ALPHA, A, LDA, X, INCX, BETA, Y, INCY)
CGEMV
Definition: cgemv.f:158
subroutine cgeru(M, N, ALPHA, X, INCX, Y, INCY, A, LDA)
CGERU
Definition: cgeru.f:130
subroutine clacgv(N, X, INCX)
CLACGV conjugates a complex vector.
Definition: clacgv.f:74
Here is the call graph for this function:
Here is the caller graph for this function: