3 * =========== DOCUMENTATION ===========
5 * Online html documentation available at
6 * http://www.netlib.org/lapack/explore-html/
9 *> Download CHERFSX + dependencies
10 *> <a href="http://www.netlib.org/cgi-bin/netlibfiles.tgz?format=tgz&filename=/lapack/lapack_routine/cherfsx.f">
12 *> <a href="http://www.netlib.org/cgi-bin/netlibfiles.zip?format=zip&filename=/lapack/lapack_routine/cherfsx.f">
14 *> <a href="http://www.netlib.org/cgi-bin/netlibfiles.txt?format=txt&filename=/lapack/lapack_routine/cherfsx.f">
21 * SUBROUTINE CHERFSX( UPLO, EQUED, N, NRHS, A, LDA, AF, LDAF, IPIV,
22 * S, B, LDB, X, LDX, RCOND, BERR, N_ERR_BNDS,
23 * ERR_BNDS_NORM, ERR_BNDS_COMP, NPARAMS, PARAMS,
26 * .. Scalar Arguments ..
27 * CHARACTER UPLO, EQUED
28 * INTEGER INFO, LDA, LDAF, LDB, LDX, N, NRHS, NPARAMS,
32 * .. Array Arguments ..
34 * COMPLEX A( LDA, * ), AF( LDAF, * ), B( LDB, * ),
35 * $ X( LDX, * ), WORK( * )
36 * REAL S( * ), PARAMS( * ), BERR( * ), RWORK( * ),
37 * $ ERR_BNDS_NORM( NRHS, * ),
38 * $ ERR_BNDS_COMP( NRHS, * )
46 *> CHERFSX improves the computed solution to a system of linear
47 *> equations when the coefficient matrix is Hermitian indefinite, and
48 *> provides error bounds and backward error estimates for the
49 *> solution. In addition to normwise error bound, the code provides
50 *> maximum componentwise error bound if possible. See comments for
51 *> ERR_BNDS_NORM and ERR_BNDS_COMP for details of the error bounds.
53 *> The original system of linear equations may have been equilibrated
54 *> before calling this routine, as described by arguments EQUED and S
55 *> below. In this case, the solution and error bounds returned are
56 *> for the original unequilibrated system.
63 *> Some optional parameters are bundled in the PARAMS array. These
64 *> settings determine how refinement is performed, but often the
65 *> defaults are acceptable. If the defaults are acceptable, users
66 *> can pass NPARAMS = 0 which prevents the source code from accessing
67 *> the PARAMS argument.
72 *> UPLO is CHARACTER*1
73 *> = 'U': Upper triangle of A is stored;
74 *> = 'L': Lower triangle of A is stored.
79 *> EQUED is CHARACTER*1
80 *> Specifies the form of equilibration that was done to A
81 *> before calling this routine. This is needed to compute
82 *> the solution and error bounds correctly.
83 *> = 'N': No equilibration
84 *> = 'Y': Both row and column equilibration, i.e., A has been
85 *> replaced by diag(S) * A * diag(S).
86 *> The right hand side B has been changed accordingly.
92 *> The order of the matrix A. N >= 0.
98 *> The number of right hand sides, i.e., the number of columns
99 *> of the matrices B and X. NRHS >= 0.
104 *> A is COMPLEX array, dimension (LDA,N)
105 *> The symmetric matrix A. If UPLO = 'U', the leading N-by-N
106 *> upper triangular part of A contains the upper triangular
107 *> part of the matrix A, and the strictly lower triangular
108 *> part of A is not referenced. If UPLO = 'L', the leading
109 *> N-by-N lower triangular part of A contains the lower
110 *> triangular part of the matrix A, and the strictly upper
111 *> triangular part of A is not referenced.
117 *> The leading dimension of the array A. LDA >= max(1,N).
122 *> AF is COMPLEX array, dimension (LDAF,N)
123 *> The factored form of the matrix A. AF contains the block
124 *> diagonal matrix D and the multipliers used to obtain the
125 *> factor U or L from the factorization A = U*D*U**T or A =
126 *> L*D*L**T as computed by SSYTRF.
132 *> The leading dimension of the array AF. LDAF >= max(1,N).
137 *> IPIV is INTEGER array, dimension (N)
138 *> Details of the interchanges and the block structure of D
139 *> as determined by SSYTRF.
144 *> S is REAL array, dimension (N)
145 *> The scale factors for A. If EQUED = 'Y', A is multiplied on
146 *> the left and right by diag(S). S is an input argument if FACT =
147 *> 'F'; otherwise, S is an output argument. If FACT = 'F' and EQUED
148 *> = 'Y', each element of S must be positive. If S is output, each
149 *> element of S is a power of the radix. If S is input, each element
150 *> of S should be a power of the radix to ensure a reliable solution
151 *> and error estimates. Scaling by powers of the radix does not cause
152 *> rounding errors unless the result underflows or overflows.
153 *> Rounding errors during scaling lead to refining with a matrix that
154 *> is not equivalent to the input matrix, producing error estimates
155 *> that may not be reliable.
160 *> B is COMPLEX array, dimension (LDB,NRHS)
161 *> The right hand side matrix B.
167 *> The leading dimension of the array B. LDB >= max(1,N).
172 *> X is COMPLEX array, dimension (LDX,NRHS)
173 *> On entry, the solution matrix X, as computed by SGETRS.
174 *> On exit, the improved solution matrix X.
180 *> The leading dimension of the array X. LDX >= max(1,N).
186 *> Reciprocal scaled condition number. This is an estimate of the
187 *> reciprocal Skeel condition number of the matrix A after
188 *> equilibration (if done). If this is less than the machine
189 *> precision (in particular, if it is zero), the matrix is singular
190 *> to working precision. Note that the error may still be small even
191 *> if this number is very small and the matrix appears ill-
197 *> BERR is REAL array, dimension (NRHS)
198 *> Componentwise relative backward error. This is the
199 *> componentwise relative backward error of each solution vector X(j)
200 *> (i.e., the smallest relative change in any element of A or B that
201 *> makes X(j) an exact solution).
204 *> \param[in] N_ERR_BNDS
206 *> N_ERR_BNDS is INTEGER
207 *> Number of error bounds to return for each right hand side
208 *> and each type (normwise or componentwise). See ERR_BNDS_NORM and
209 *> ERR_BNDS_COMP below.
212 *> \param[out] ERR_BNDS_NORM
214 *> ERR_BNDS_NORM is REAL array, dimension (NRHS, N_ERR_BNDS)
215 *> For each right-hand side, this array contains information about
216 *> various error bounds and condition numbers corresponding to the
217 *> normwise relative error, which is defined as follows:
219 *> Normwise relative error in the ith solution vector:
220 *> max_j (abs(XTRUE(j,i) - X(j,i)))
221 *> ------------------------------
224 *> The array is indexed by the type of error information as described
225 *> below. There currently are up to three pieces of information
228 *> The first index in ERR_BNDS_NORM(i,:) corresponds to the ith
231 *> The second index in ERR_BNDS_NORM(:,err) contains the following
233 *> err = 1 "Trust/don't trust" boolean. Trust the answer if the
234 *> reciprocal condition number is less than the threshold
235 *> sqrt(n) * slamch('Epsilon').
237 *> err = 2 "Guaranteed" error bound: The estimated forward error,
238 *> almost certainly within a factor of 10 of the true error
239 *> so long as the next entry is greater than the threshold
240 *> sqrt(n) * slamch('Epsilon'). This error bound should only
241 *> be trusted if the previous boolean is true.
243 *> err = 3 Reciprocal condition number: Estimated normwise
244 *> reciprocal condition number. Compared with the threshold
245 *> sqrt(n) * slamch('Epsilon') to determine if the error
246 *> estimate is "guaranteed". These reciprocal condition
247 *> numbers are 1 / (norm(Z^{-1},inf) * norm(Z,inf)) for some
248 *> appropriately scaled matrix Z.
249 *> Let Z = S*A, where S scales each row by a power of the
250 *> radix so all absolute row sums of Z are approximately 1.
252 *> See Lapack Working Note 165 for further details and extra
256 *> \param[out] ERR_BNDS_COMP
258 *> ERR_BNDS_COMP is REAL array, dimension (NRHS, N_ERR_BNDS)
259 *> For each right-hand side, this array contains information about
260 *> various error bounds and condition numbers corresponding to the
261 *> componentwise relative error, which is defined as follows:
263 *> Componentwise relative error in the ith solution vector:
264 *> abs(XTRUE(j,i) - X(j,i))
265 *> max_j ----------------------
268 *> The array is indexed by the right-hand side i (on which the
269 *> componentwise relative error depends), and the type of error
270 *> information as described below. There currently are up to three
271 *> pieces of information returned for each right-hand side. If
272 *> componentwise accuracy is not requested (PARAMS(3) = 0.0), then
273 *> ERR_BNDS_COMP is not accessed. If N_ERR_BNDS .LT. 3, then at most
274 *> the first (:,N_ERR_BNDS) entries are returned.
276 *> The first index in ERR_BNDS_COMP(i,:) corresponds to the ith
279 *> The second index in ERR_BNDS_COMP(:,err) contains the following
281 *> err = 1 "Trust/don't trust" boolean. Trust the answer if the
282 *> reciprocal condition number is less than the threshold
283 *> sqrt(n) * slamch('Epsilon').
285 *> err = 2 "Guaranteed" error bound: The estimated forward error,
286 *> almost certainly within a factor of 10 of the true error
287 *> so long as the next entry is greater than the threshold
288 *> sqrt(n) * slamch('Epsilon'). This error bound should only
289 *> be trusted if the previous boolean is true.
291 *> err = 3 Reciprocal condition number: Estimated componentwise
292 *> reciprocal condition number. Compared with the threshold
293 *> sqrt(n) * slamch('Epsilon') to determine if the error
294 *> estimate is "guaranteed". These reciprocal condition
295 *> numbers are 1 / (norm(Z^{-1},inf) * norm(Z,inf)) for some
296 *> appropriately scaled matrix Z.
297 *> Let Z = S*(A*diag(x)), where x is the solution for the
298 *> current right-hand side and S scales each row of
299 *> A*diag(x) by a power of the radix so all absolute row
300 *> sums of Z are approximately 1.
302 *> See Lapack Working Note 165 for further details and extra
306 *> \param[in] NPARAMS
308 *> NPARAMS is INTEGER
309 *> Specifies the number of parameters set in PARAMS. If .LE. 0, the
310 *> PARAMS array is never referenced and default values are used.
313 *> \param[in,out] PARAMS
315 *> PARAMS is REAL array, dimension NPARAMS
316 *> Specifies algorithm parameters. If an entry is .LT. 0.0, then
317 *> that entry will be filled with default value used for that
318 *> parameter. Only positions up to NPARAMS are accessed; defaults
319 *> are used for higher-numbered parameters.
321 *> PARAMS(LA_LINRX_ITREF_I = 1) : Whether to perform iterative
322 *> refinement or not.
324 *> = 0.0 : No refinement is performed, and no error bounds are
326 *> = 1.0 : Use the double-precision refinement algorithm,
327 *> possibly with doubled-single computations if the
328 *> compilation environment does not support DOUBLE
330 *> (other values are reserved for future use)
332 *> PARAMS(LA_LINRX_ITHRESH_I = 2) : Maximum number of residual
333 *> computations allowed for refinement.
335 *> Aggressive: Set to 100 to permit convergence using approximate
336 *> factorizations or factorizations other than LU. If
337 *> the factorization uses a technique other than
338 *> Gaussian elimination, the guarantees in
339 *> err_bnds_norm and err_bnds_comp may no longer be
342 *> PARAMS(LA_LINRX_CWISE_I = 3) : Flag determining if the code
343 *> will attempt to find a solution with small componentwise
344 *> relative error in the double-precision algorithm. Positive
345 *> is true, 0.0 is false.
346 *> Default: 1.0 (attempt componentwise convergence)
351 *> WORK is COMPLEX array, dimension (2*N)
356 *> RWORK is REAL array, dimension (2*N)
362 *> = 0: Successful exit. The solution to every right-hand side is
364 *> < 0: If INFO = -i, the i-th argument had an illegal value
365 *> > 0 and <= N: U(INFO,INFO) is exactly zero. The factorization
366 *> has been completed, but the factor U is exactly singular, so
367 *> the solution and error bounds could not be computed. RCOND = 0
369 *> = N+J: The solution corresponding to the Jth right-hand side is
370 *> not guaranteed. The solutions corresponding to other right-
371 *> hand sides K with K > J may not be guaranteed as well, but
372 *> only the first such right-hand side is reported. If a small
373 *> componentwise error is not requested (PARAMS(3) = 0.0) then
374 *> the Jth right-hand side is the first with a normwise error
375 *> bound that is not guaranteed (the smallest J such
376 *> that ERR_BNDS_NORM(J,1) = 0.0). By default (PARAMS(3) = 1.0)
377 *> the Jth right-hand side is the first with either a normwise or
378 *> componentwise error bound that is not guaranteed (the smallest
379 *> J such that either ERR_BNDS_NORM(J,1) = 0.0 or
380 *> ERR_BNDS_COMP(J,1) = 0.0). See the definition of
381 *> ERR_BNDS_NORM(:,1) and ERR_BNDS_COMP(:,1). To get information
382 *> about all of the right-hand sides check ERR_BNDS_NORM or
389 *> \author Univ. of Tennessee
390 *> \author Univ. of California Berkeley
391 *> \author Univ. of Colorado Denver
396 *> \ingroup complexHEcomputational
398 * =====================================================================
399 SUBROUTINE CHERFSX( UPLO, EQUED, N, NRHS, A, LDA, AF, LDAF, IPIV,
400 $ S, B, LDB, X, LDX, RCOND, BERR, N_ERR_BNDS,
401 $ ERR_BNDS_NORM, ERR_BNDS_COMP, NPARAMS, PARAMS,
402 $ WORK, RWORK, INFO )
404 * -- LAPACK computational routine (version 3.4.1) --
405 * -- LAPACK is a software package provided by Univ. of Tennessee, --
406 * -- Univ. of California Berkeley, Univ. of Colorado Denver and NAG Ltd..--
409 * .. Scalar Arguments ..
410 CHARACTER UPLO, EQUED
411 INTEGER INFO, LDA, LDAF, LDB, LDX, N, NRHS, NPARAMS,
415 * .. Array Arguments ..
417 COMPLEX A( LDA, * ), AF( LDAF, * ), B( LDB, * ),
418 $ X( LDX, * ), WORK( * )
419 REAL S( * ), PARAMS( * ), BERR( * ), RWORK( * ),
420 $ ERR_BNDS_NORM( NRHS, * ),
421 $ ERR_BNDS_COMP( NRHS, * )
423 * ==================================================================
427 PARAMETER ( ZERO = 0.0E+0, ONE = 1.0E+0 )
428 REAL ITREF_DEFAULT, ITHRESH_DEFAULT,
429 $ COMPONENTWISE_DEFAULT
430 REAL RTHRESH_DEFAULT, DZTHRESH_DEFAULT
431 PARAMETER ( ITREF_DEFAULT = 1.0 )
432 PARAMETER ( ITHRESH_DEFAULT = 10.0 )
433 PARAMETER ( COMPONENTWISE_DEFAULT = 1.0 )
434 PARAMETER ( RTHRESH_DEFAULT = 0.5 )
435 PARAMETER ( DZTHRESH_DEFAULT = 0.25 )
436 INTEGER LA_LINRX_ITREF_I, LA_LINRX_ITHRESH_I,
438 PARAMETER ( LA_LINRX_ITREF_I = 1,
439 $ LA_LINRX_ITHRESH_I = 2 )
440 PARAMETER ( LA_LINRX_CWISE_I = 3 )
441 INTEGER LA_LINRX_TRUST_I, LA_LINRX_ERR_I,
443 PARAMETER ( LA_LINRX_TRUST_I = 1, LA_LINRX_ERR_I = 2 )
444 PARAMETER ( LA_LINRX_RCOND_I = 3 )
446 * .. Local Scalars ..
449 INTEGER J, PREC_TYPE, REF_TYPE
451 REAL ANORM, RCOND_TMP
452 REAL ILLRCOND_THRESH, ERR_LBND, CWISE_WRONG
455 REAL RTHRESH, UNSTABLE_THRESH
457 * .. External Subroutines ..
458 EXTERNAL XERBLA, CHECON, CLA_HERFSX_EXTENDED
460 * .. Intrinsic Functions ..
461 INTRINSIC MAX, SQRT, TRANSFER
463 * .. External Functions ..
464 EXTERNAL LSAME, BLAS_FPINFO_X, ILATRANS, ILAPREC
465 EXTERNAL SLAMCH, CLANHE, CLA_HERCOND_X, CLA_HERCOND_C
466 REAL SLAMCH, CLANHE, CLA_HERCOND_X, CLA_HERCOND_C
468 INTEGER BLAS_FPINFO_X
469 INTEGER ILATRANS, ILAPREC
471 * .. Executable Statements ..
473 * Check the input parameters.
476 REF_TYPE = INT( ITREF_DEFAULT )
477 IF ( NPARAMS .GE. LA_LINRX_ITREF_I ) THEN
478 IF ( PARAMS( LA_LINRX_ITREF_I ) .LT. 0.0 ) THEN
479 PARAMS( LA_LINRX_ITREF_I ) = ITREF_DEFAULT
481 REF_TYPE = PARAMS( LA_LINRX_ITREF_I )
485 * Set default parameters.
487 ILLRCOND_THRESH = REAL( N ) * SLAMCH( 'Epsilon' )
488 ITHRESH = INT( ITHRESH_DEFAULT )
489 RTHRESH = RTHRESH_DEFAULT
490 UNSTABLE_THRESH = DZTHRESH_DEFAULT
491 IGNORE_CWISE = COMPONENTWISE_DEFAULT .EQ. 0.0
493 IF ( NPARAMS.GE.LA_LINRX_ITHRESH_I ) THEN
494 IF ( PARAMS( LA_LINRX_ITHRESH_I ).LT.0.0 ) THEN
495 PARAMS( LA_LINRX_ITHRESH_I ) = ITHRESH
497 ITHRESH = INT( PARAMS( LA_LINRX_ITHRESH_I ) )
500 IF ( NPARAMS.GE.LA_LINRX_CWISE_I ) THEN
501 IF ( PARAMS(LA_LINRX_CWISE_I ).LT.0.0 ) THEN
502 IF ( IGNORE_CWISE ) THEN
503 PARAMS( LA_LINRX_CWISE_I ) = 0.0
505 PARAMS( LA_LINRX_CWISE_I ) = 1.0
508 IGNORE_CWISE = PARAMS( LA_LINRX_CWISE_I ) .EQ. 0.0
511 IF ( REF_TYPE .EQ. 0 .OR. N_ERR_BNDS .EQ. 0 ) THEN
513 ELSE IF ( IGNORE_CWISE ) THEN
519 RCEQU = LSAME( EQUED, 'Y' )
521 * Test input parameters.
523 IF (.NOT.LSAME( UPLO, 'U' ) .AND. .NOT.LSAME( UPLO, 'L' ) ) THEN
525 ELSE IF( .NOT.RCEQU .AND. .NOT.LSAME( EQUED, 'N' ) ) THEN
527 ELSE IF( N.LT.0 ) THEN
529 ELSE IF( NRHS.LT.0 ) THEN
531 ELSE IF( LDA.LT.MAX( 1, N ) ) THEN
533 ELSE IF( LDAF.LT.MAX( 1, N ) ) THEN
535 ELSE IF( LDB.LT.MAX( 1, N ) ) THEN
537 ELSE IF( LDX.LT.MAX( 1, N ) ) THEN
541 CALL XERBLA( 'CHERFSX', -INFO )
545 * Quick return if possible.
547 IF( N.EQ.0 .OR. NRHS.EQ.0 ) THEN
551 IF ( N_ERR_BNDS .GE. 1 ) THEN
552 ERR_BNDS_NORM( J, LA_LINRX_TRUST_I ) = 1.0
553 ERR_BNDS_COMP( J, LA_LINRX_TRUST_I ) = 1.0
555 IF ( N_ERR_BNDS .GE. 2 ) THEN
556 ERR_BNDS_NORM( J, LA_LINRX_ERR_I ) = 0.0
557 ERR_BNDS_COMP( J, LA_LINRX_ERR_I ) = 0.0
559 IF ( N_ERR_BNDS .GE. 3 ) THEN
560 ERR_BNDS_NORM( J, LA_LINRX_RCOND_I ) = 1.0
561 ERR_BNDS_COMP( J, LA_LINRX_RCOND_I ) = 1.0
567 * Default to failure.
572 IF ( N_ERR_BNDS .GE. 1 ) THEN
573 ERR_BNDS_NORM( J, LA_LINRX_TRUST_I ) = 1.0
574 ERR_BNDS_COMP( J, LA_LINRX_TRUST_I ) = 1.0
576 IF ( N_ERR_BNDS .GE. 2 ) THEN
577 ERR_BNDS_NORM( J, LA_LINRX_ERR_I ) = 1.0
578 ERR_BNDS_COMP( J, LA_LINRX_ERR_I ) = 1.0
580 IF ( N_ERR_BNDS .GE. 3 ) THEN
581 ERR_BNDS_NORM( J, LA_LINRX_RCOND_I ) = 0.0
582 ERR_BNDS_COMP( J, LA_LINRX_RCOND_I ) = 0.0
586 * Compute the norm of A and the reciprocal of the condition
590 ANORM = CLANHE( NORM, UPLO, N, A, LDA, RWORK )
591 CALL CHECON( UPLO, N, AF, LDAF, IPIV, ANORM, RCOND, WORK,
594 * Perform refinement on each right-hand side
596 IF ( REF_TYPE .NE. 0 ) THEN
598 PREC_TYPE = ILAPREC( 'D' )
600 CALL CLA_HERFSX_EXTENDED( PREC_TYPE, UPLO, N,
601 $ NRHS, A, LDA, AF, LDAF, IPIV, RCEQU, S, B,
602 $ LDB, X, LDX, BERR, N_NORMS, ERR_BNDS_NORM, ERR_BNDS_COMP,
603 $ WORK, RWORK, WORK(N+1),
604 $ TRANSFER (RWORK(1:2*N), (/ (ZERO, ZERO) /), N), RCOND,
605 $ ITHRESH, RTHRESH, UNSTABLE_THRESH, IGNORE_CWISE,
609 ERR_LBND = MAX( 10.0, SQRT( REAL( N ) ) ) * SLAMCH( 'Epsilon' )
610 IF ( N_ERR_BNDS .GE. 1 .AND. N_NORMS .GE. 1 ) THEN
612 * Compute scaled normwise condition number cond(A*C).
615 RCOND_TMP = CLA_HERCOND_C( UPLO, N, A, LDA, AF, LDAF, IPIV,
616 $ S, .TRUE., INFO, WORK, RWORK )
618 RCOND_TMP = CLA_HERCOND_C( UPLO, N, A, LDA, AF, LDAF, IPIV,
619 $ S, .FALSE., INFO, WORK, RWORK )
623 * Cap the error at 1.0.
625 IF ( N_ERR_BNDS .GE. LA_LINRX_ERR_I
626 $ .AND. ERR_BNDS_NORM( J, LA_LINRX_ERR_I ) .GT. 1.0 )
627 $ ERR_BNDS_NORM( J, LA_LINRX_ERR_I ) = 1.0
629 * Threshold the error (see LAWN).
631 IF (RCOND_TMP .LT. ILLRCOND_THRESH) THEN
632 ERR_BNDS_NORM( J, LA_LINRX_ERR_I ) = 1.0
633 ERR_BNDS_NORM( J, LA_LINRX_TRUST_I ) = 0.0
634 IF ( INFO .LE. N ) INFO = N + J
635 ELSE IF ( ERR_BNDS_NORM( J, LA_LINRX_ERR_I ) .LT. ERR_LBND )
637 ERR_BNDS_NORM( J, LA_LINRX_ERR_I ) = ERR_LBND
638 ERR_BNDS_NORM( J, LA_LINRX_TRUST_I ) = 1.0
641 * Save the condition number.
643 IF ( N_ERR_BNDS .GE. LA_LINRX_RCOND_I ) THEN
644 ERR_BNDS_NORM( J, LA_LINRX_RCOND_I ) = RCOND_TMP
649 IF ( N_ERR_BNDS .GE. 1 .AND. N_NORMS .GE. 2 ) THEN
651 * Compute componentwise condition number cond(A*diag(Y(:,J))) for
652 * each right-hand side using the current solution as an estimate of
653 * the true solution. If the componentwise error estimate is too
654 * large, then the solution is a lousy estimate of truth and the
655 * estimated RCOND may be too optimistic. To avoid misleading users,
656 * the inverse condition number is set to 0.0 when the estimated
657 * cwise error is at least CWISE_WRONG.
659 CWISE_WRONG = SQRT( SLAMCH( 'Epsilon' ) )
661 IF ( ERR_BNDS_COMP( J, LA_LINRX_ERR_I ) .LT. CWISE_WRONG )
663 RCOND_TMP = CLA_HERCOND_X( UPLO, N, A, LDA, AF, LDAF,
664 $ IPIV, X( 1, J ), INFO, WORK, RWORK )
669 * Cap the error at 1.0.
671 IF ( N_ERR_BNDS .GE. LA_LINRX_ERR_I
672 $ .AND. ERR_BNDS_COMP( J, LA_LINRX_ERR_I ) .GT. 1.0 )
673 $ ERR_BNDS_COMP( J, LA_LINRX_ERR_I ) = 1.0
675 * Threshold the error (see LAWN).
677 IF ( RCOND_TMP .LT. ILLRCOND_THRESH ) THEN
678 ERR_BNDS_COMP( J, LA_LINRX_ERR_I ) = 1.0
679 ERR_BNDS_COMP( J, LA_LINRX_TRUST_I ) = 0.0
680 IF ( .NOT. IGNORE_CWISE
681 $ .AND. INFO.LT.N + J ) INFO = N + J
682 ELSE IF ( ERR_BNDS_COMP( J, LA_LINRX_ERR_I )
683 $ .LT. ERR_LBND ) THEN
684 ERR_BNDS_COMP( J, LA_LINRX_ERR_I ) = ERR_LBND
685 ERR_BNDS_COMP( J, LA_LINRX_TRUST_I ) = 1.0
688 * Save the condition number.
690 IF ( N_ERR_BNDS .GE. LA_LINRX_RCOND_I ) THEN
691 ERR_BNDS_COMP( J, LA_LINRX_RCOND_I ) = RCOND_TMP