SUBROUTINE DLAEBZ( IJOB, NITMAX, N, MMAX, MINP, NBMIN, ABSTOL, $ RELTOL, PIVMIN, D, E, E2, NVAL, AB, C, MOUT, $ NAB, WORK, IWORK, INFO ) * * -- LAPACK auxiliary routine (version 3.1) -- * Univ. of Tennessee, Univ. of California Berkeley and NAG Ltd.. * November 2006 * * .. Scalar Arguments .. INTEGER IJOB, INFO, MINP, MMAX, MOUT, N, NBMIN, NITMAX DOUBLE PRECISION ABSTOL, PIVMIN, RELTOL * .. * .. Array Arguments .. INTEGER IWORK( * ), NAB( MMAX, * ), NVAL( * ) DOUBLE PRECISION AB( MMAX, * ), C( * ), D( * ), E( * ), E2( * ), $ WORK( * ) * .. * * Purpose * ======= * * DLAEBZ contains the iteration loops which compute and use the * function N(w), which is the count of eigenvalues of a symmetric * tridiagonal matrix T less than or equal to its argument w. It * performs a choice of two types of loops: * * IJOB=1, followed by * IJOB=2: It takes as input a list of intervals and returns a list of * sufficiently small intervals whose union contains the same * eigenvalues as the union of the original intervals. * The input intervals are (AB(j,1),AB(j,2)], j=1,...,MINP. * The output interval (AB(j,1),AB(j,2)] will contain * eigenvalues NAB(j,1)+1,...,NAB(j,2), where 1 <= j <= MOUT. * * IJOB=3: It performs a binary search in each input interval * (AB(j,1),AB(j,2)] for a point w(j) such that * N(w(j))=NVAL(j), and uses C(j) as the starting point of * the search. If such a w(j) is found, then on output * AB(j,1)=AB(j,2)=w. If no such w(j) is found, then on output * (AB(j,1),AB(j,2)] will be a small interval containing the * point where N(w) jumps through NVAL(j), unless that point * lies outside the initial interval. * * Note that the intervals are in all cases half-open intervals, * i.e., of the form (a,b] , which includes b but not a . * * To avoid underflow, the matrix should be scaled so that its largest * element is no greater than overflow**(1/2) * underflow**(1/4) * in absolute value. To assure the most accurate computation * of small eigenvalues, the matrix should be scaled to be * not much smaller than that, either. * * See W. Kahan "Accurate Eigenvalues of a Symmetric Tridiagonal * Matrix", Report CS41, Computer Science Dept., Stanford * University, July 21, 1966 * * Note: the arguments are, in general, *not* checked for unreasonable * values. * * Arguments * ========= * * IJOB (input) INTEGER * Specifies what is to be done: * = 1: Compute NAB for the initial intervals. * = 2: Perform bisection iteration to find eigenvalues of T. * = 3: Perform bisection iteration to invert N(w), i.e., * to find a point which has a specified number of * eigenvalues of T to its left. * Other values will cause DLAEBZ to return with INFO=-1. * * NITMAX (input) INTEGER * The maximum number of "levels" of bisection to be * performed, i.e., an interval of width W will not be made * smaller than 2^(-NITMAX) * W. If not all intervals * have converged after NITMAX iterations, then INFO is set * to the number of non-converged intervals. * * N (input) INTEGER * The dimension n of the tridiagonal matrix T. It must be at * least 1. * * MMAX (input) INTEGER * The maximum number of intervals. If more than MMAX intervals * are generated, then DLAEBZ will quit with INFO=MMAX+1. * * MINP (input) INTEGER * The initial number of intervals. It may not be greater than * MMAX. * * NBMIN (input) INTEGER * The smallest number of intervals that should be processed * using a vector loop. If zero, then only the scalar loop * will be used. * * ABSTOL (input) DOUBLE PRECISION * The minimum (absolute) width of an interval. When an * interval is narrower than ABSTOL, or than RELTOL times the * larger (in magnitude) endpoint, then it is considered to be * sufficiently small, i.e., converged. This must be at least * zero. * * RELTOL (input) DOUBLE PRECISION * The minimum relative width of an interval. When an interval * is narrower than ABSTOL, or than RELTOL times the larger (in * magnitude) endpoint, then it is considered to be * sufficiently small, i.e., converged. Note: this should * always be at least radix*machine epsilon. * * PIVMIN (input) DOUBLE PRECISION * The minimum absolute value of a "pivot" in the Sturm * sequence loop. This *must* be at least max |e(j)**2| * * safe_min and at least safe_min, where safe_min is at least * the smallest number that can divide one without overflow. * * D (input) DOUBLE PRECISION array, dimension (N) * The diagonal elements of the tridiagonal matrix T. * * E (input) DOUBLE PRECISION array, dimension (N) * The offdiagonal elements of the tridiagonal matrix T in * positions 1 through N-1. E(N) is arbitrary. * * E2 (input) DOUBLE PRECISION array, dimension (N) * The squares of the offdiagonal elements of the tridiagonal * matrix T. E2(N) is ignored. * * NVAL (input/output) INTEGER array, dimension (MINP) * If IJOB=1 or 2, not referenced. * If IJOB=3, the desired values of N(w). The elements of NVAL * will be reordered to correspond with the intervals in AB. * Thus, NVAL(j) on output will not, in general be the same as * NVAL(j) on input, but it will correspond with the interval * (AB(j,1),AB(j,2)] on output. * * AB (input/output) DOUBLE PRECISION array, dimension (MMAX,2) * The endpoints of the intervals. AB(j,1) is a(j), the left * endpoint of the j-th interval, and AB(j,2) is b(j), the * right endpoint of the j-th interval. The input intervals * will, in general, be modified, split, and reordered by the * calculation. * * C (input/output) DOUBLE PRECISION array, dimension (MMAX) * If IJOB=1, ignored. * If IJOB=2, workspace. * If IJOB=3, then on input C(j) should be initialized to the * first search point in the binary search. * * MOUT (output) INTEGER * If IJOB=1, the number of eigenvalues in the intervals. * If IJOB=2 or 3, the number of intervals output. * If IJOB=3, MOUT will equal MINP. * * NAB (input/output) INTEGER array, dimension (MMAX,2) * If IJOB=1, then on output NAB(i,j) will be set to N(AB(i,j)). * If IJOB=2, then on input, NAB(i,j) should be set. It must * satisfy the condition: * N(AB(i,1)) <= NAB(i,1) <= NAB(i,2) <= N(AB(i,2)), * which means that in interval i only eigenvalues * NAB(i,1)+1,...,NAB(i,2) will be considered. Usually, * NAB(i,j)=N(AB(i,j)), from a previous call to DLAEBZ with * IJOB=1. * On output, NAB(i,j) will contain * max(na(k),min(nb(k),N(AB(i,j)))), where k is the index of * the input interval that the output interval * (AB(j,1),AB(j,2)] came from, and na(k) and nb(k) are the * the input values of NAB(k,1) and NAB(k,2). * If IJOB=3, then on output, NAB(i,j) contains N(AB(i,j)), * unless N(w) > NVAL(i) for all search points w , in which * case NAB(i,1) will not be modified, i.e., the output * value will be the same as the input value (modulo * reorderings -- see NVAL and AB), or unless N(w) < NVAL(i) * for all search points w , in which case NAB(i,2) will * not be modified. Normally, NAB should be set to some * distinctive value(s) before DLAEBZ is called. * * WORK (workspace) DOUBLE PRECISION array, dimension (MMAX) * Workspace. * * IWORK (workspace) INTEGER array, dimension (MMAX) * Workspace. * * INFO (output) INTEGER * = 0: All intervals converged. * = 1--MMAX: The last INFO intervals did not converge. * = MMAX+1: More than MMAX intervals were generated. * * Further Details * =============== * * This routine is intended to be called only by other LAPACK * routines, thus the interface is less user-friendly. It is intended * for two purposes: * * (a) finding eigenvalues. In this case, DLAEBZ should have one or * more initial intervals set up in AB, and DLAEBZ should be called * with IJOB=1. This sets up NAB, and also counts the eigenvalues. * Intervals with no eigenvalues would usually be thrown out at * this point. Also, if not all the eigenvalues in an interval i * are desired, NAB(i,1) can be increased or NAB(i,2) decreased. * For example, set NAB(i,1)=NAB(i,2)-1 to get the largest * eigenvalue. DLAEBZ is then called with IJOB=2 and MMAX * no smaller than the value of MOUT returned by the call with * IJOB=1. After this (IJOB=2) call, eigenvalues NAB(i,1)+1 * through NAB(i,2) are approximately AB(i,1) (or AB(i,2)) to the * tolerance specified by ABSTOL and RELTOL. * * (b) finding an interval (a',b'] containing eigenvalues w(f),...,w(l). * In this case, start with a Gershgorin interval (a,b). Set up * AB to contain 2 search intervals, both initially (a,b). One * NVAL element should contain f-1 and the other should contain l * , while C should contain a and b, resp. NAB(i,1) should be -1 * and NAB(i,2) should be N+1, to flag an error if the desired * interval does not lie in (a,b). DLAEBZ is then called with * IJOB=3. On exit, if w(f-1) < w(f), then one of the intervals -- * j -- will have AB(j,1)=AB(j,2) and NAB(j,1)=NAB(j,2)=f-1, while * if, to the specified tolerance, w(f-k)=...=w(f+r), k > 0 and r * >= 0, then the interval will have N(AB(j,1))=NAB(j,1)=f-k and * N(AB(j,2))=NAB(j,2)=f+r. The cases w(l) < w(l+1) and * w(l-r)=...=w(l+k) are handled similarly. * * ===================================================================== * * .. Parameters .. DOUBLE PRECISION ZERO, TWO, HALF PARAMETER ( ZERO = 0.0D0, TWO = 2.0D0, $ HALF = 1.0D0 / TWO ) * .. * .. Local Scalars .. INTEGER ITMP1, ITMP2, J, JI, JIT, JP, KF, KFNEW, KL, $ KLNEW DOUBLE PRECISION TMP1, TMP2 * .. * .. Intrinsic Functions .. INTRINSIC ABS, MAX, MIN * .. * .. Executable Statements .. * * Check for Errors * INFO = 0 IF( IJOB.LT.1 .OR. IJOB.GT.3 ) THEN INFO = -1 RETURN END IF * * Initialize NAB * IF( IJOB.EQ.1 ) THEN * * Compute the number of eigenvalues in the initial intervals. * MOUT = 0 *DIR$ NOVECTOR DO 30 JI = 1, MINP DO 20 JP = 1, 2 TMP1 = D( 1 ) - AB( JI, JP ) IF( ABS( TMP1 ).LT.PIVMIN ) $ TMP1 = -PIVMIN NAB( JI, JP ) = 0 IF( TMP1.LE.ZERO ) $ NAB( JI, JP ) = 1 * DO 10 J = 2, N TMP1 = D( J ) - E2( J-1 ) / TMP1 - AB( JI, JP ) IF( ABS( TMP1 ).LT.PIVMIN ) $ TMP1 = -PIVMIN IF( TMP1.LE.ZERO ) $ NAB( JI, JP ) = NAB( JI, JP ) + 1 10 CONTINUE 20 CONTINUE MOUT = MOUT + NAB( JI, 2 ) - NAB( JI, 1 ) 30 CONTINUE RETURN END IF * * Initialize for loop * * KF and KL have the following meaning: * Intervals 1,...,KF-1 have converged. * Intervals KF,...,KL still need to be refined. * KF = 1 KL = MINP * * If IJOB=2, initialize C. * If IJOB=3, use the user-supplied starting point. * IF( IJOB.EQ.2 ) THEN DO 40 JI = 1, MINP C( JI ) = HALF*( AB( JI, 1 )+AB( JI, 2 ) ) 40 CONTINUE END IF * * Iteration loop * DO 130 JIT = 1, NITMAX * * Loop over intervals * IF( KL-KF+1.GE.NBMIN .AND. NBMIN.GT.0 ) THEN * * Begin of Parallel Version of the loop * DO 60 JI = KF, KL * * Compute N(c), the number of eigenvalues less than c * WORK( JI ) = D( 1 ) - C( JI ) IWORK( JI ) = 0 IF( WORK( JI ).LE.PIVMIN ) THEN IWORK( JI ) = 1 WORK( JI ) = MIN( WORK( JI ), -PIVMIN ) END IF * DO 50 J = 2, N WORK( JI ) = D( J ) - E2( J-1 ) / WORK( JI ) - C( JI ) IF( WORK( JI ).LE.PIVMIN ) THEN IWORK( JI ) = IWORK( JI ) + 1 WORK( JI ) = MIN( WORK( JI ), -PIVMIN ) END IF 50 CONTINUE 60 CONTINUE * IF( IJOB.LE.2 ) THEN * * IJOB=2: Choose all intervals containing eigenvalues. * KLNEW = KL DO 70 JI = KF, KL * * Insure that N(w) is monotone * IWORK( JI ) = MIN( NAB( JI, 2 ), $ MAX( NAB( JI, 1 ), IWORK( JI ) ) ) * * Update the Queue -- add intervals if both halves * contain eigenvalues. * IF( IWORK( JI ).EQ.NAB( JI, 2 ) ) THEN * * No eigenvalue in the upper interval: * just use the lower interval. * AB( JI, 2 ) = C( JI ) * ELSE IF( IWORK( JI ).EQ.NAB( JI, 1 ) ) THEN * * No eigenvalue in the lower interval: * just use the upper interval. * AB( JI, 1 ) = C( JI ) ELSE KLNEW = KLNEW + 1 IF( KLNEW.LE.MMAX ) THEN * * Eigenvalue in both intervals -- add upper to * queue. * AB( KLNEW, 2 ) = AB( JI, 2 ) NAB( KLNEW, 2 ) = NAB( JI, 2 ) AB( KLNEW, 1 ) = C( JI ) NAB( KLNEW, 1 ) = IWORK( JI ) AB( JI, 2 ) = C( JI ) NAB( JI, 2 ) = IWORK( JI ) ELSE INFO = MMAX + 1 END IF END IF 70 CONTINUE IF( INFO.NE.0 ) $ RETURN KL = KLNEW ELSE * * IJOB=3: Binary search. Keep only the interval containing * w s.t. N(w) = NVAL * DO 80 JI = KF, KL IF( IWORK( JI ).LE.NVAL( JI ) ) THEN AB( JI, 1 ) = C( JI ) NAB( JI, 1 ) = IWORK( JI ) END IF IF( IWORK( JI ).GE.NVAL( JI ) ) THEN AB( JI, 2 ) = C( JI ) NAB( JI, 2 ) = IWORK( JI ) END IF 80 CONTINUE END IF * ELSE * * End of Parallel Version of the loop * * Begin of Serial Version of the loop * KLNEW = KL DO 100 JI = KF, KL * * Compute N(w), the number of eigenvalues less than w * TMP1 = C( JI ) TMP2 = D( 1 ) - TMP1 ITMP1 = 0 IF( TMP2.LE.PIVMIN ) THEN ITMP1 = 1 TMP2 = MIN( TMP2, -PIVMIN ) END IF * * A series of compiler directives to defeat vectorization * for the next loop * *$PL$ CMCHAR=' ' CDIR$ NEXTSCALAR C$DIR SCALAR CDIR$ NEXT SCALAR CVD$L NOVECTOR CDEC$ NOVECTOR CVD$ NOVECTOR *VDIR NOVECTOR *VOCL LOOP,SCALAR CIBM PREFER SCALAR *$PL$ CMCHAR='*' * DO 90 J = 2, N TMP2 = D( J ) - E2( J-1 ) / TMP2 - TMP1 IF( TMP2.LE.PIVMIN ) THEN ITMP1 = ITMP1 + 1 TMP2 = MIN( TMP2, -PIVMIN ) END IF 90 CONTINUE * IF( IJOB.LE.2 ) THEN * * IJOB=2: Choose all intervals containing eigenvalues. * * Insure that N(w) is monotone * ITMP1 = MIN( NAB( JI, 2 ), $ MAX( NAB( JI, 1 ), ITMP1 ) ) * * Update the Queue -- add intervals if both halves * contain eigenvalues. * IF( ITMP1.EQ.NAB( JI, 2 ) ) THEN * * No eigenvalue in the upper interval: * just use the lower interval. * AB( JI, 2 ) = TMP1 * ELSE IF( ITMP1.EQ.NAB( JI, 1 ) ) THEN * * No eigenvalue in the lower interval: * just use the upper interval. * AB( JI, 1 ) = TMP1 ELSE IF( KLNEW.LT.MMAX ) THEN * * Eigenvalue in both intervals -- add upper to queue. * KLNEW = KLNEW + 1 AB( KLNEW, 2 ) = AB( JI, 2 ) NAB( KLNEW, 2 ) = NAB( JI, 2 ) AB( KLNEW, 1 ) = TMP1 NAB( KLNEW, 1 ) = ITMP1 AB( JI, 2 ) = TMP1 NAB( JI, 2 ) = ITMP1 ELSE INFO = MMAX + 1 RETURN END IF ELSE * * IJOB=3: Binary search. Keep only the interval * containing w s.t. N(w) = NVAL * IF( ITMP1.LE.NVAL( JI ) ) THEN AB( JI, 1 ) = TMP1 NAB( JI, 1 ) = ITMP1 END IF IF( ITMP1.GE.NVAL( JI ) ) THEN AB( JI, 2 ) = TMP1 NAB( JI, 2 ) = ITMP1 END IF END IF 100 CONTINUE KL = KLNEW * * End of Serial Version of the loop * END IF * * Check for convergence * KFNEW = KF DO 110 JI = KF, KL TMP1 = ABS( AB( JI, 2 )-AB( JI, 1 ) ) TMP2 = MAX( ABS( AB( JI, 2 ) ), ABS( AB( JI, 1 ) ) ) IF( TMP1.LT.MAX( ABSTOL, PIVMIN, RELTOL*TMP2 ) .OR. $ NAB( JI, 1 ).GE.NAB( JI, 2 ) ) THEN * * Converged -- Swap with position KFNEW, * then increment KFNEW * IF( JI.GT.KFNEW ) THEN TMP1 = AB( JI, 1 ) TMP2 = AB( JI, 2 ) ITMP1 = NAB( JI, 1 ) ITMP2 = NAB( JI, 2 ) AB( JI, 1 ) = AB( KFNEW, 1 ) AB( JI, 2 ) = AB( KFNEW, 2 ) NAB( JI, 1 ) = NAB( KFNEW, 1 ) NAB( JI, 2 ) = NAB( KFNEW, 2 ) AB( KFNEW, 1 ) = TMP1 AB( KFNEW, 2 ) = TMP2 NAB( KFNEW, 1 ) = ITMP1 NAB( KFNEW, 2 ) = ITMP2 IF( IJOB.EQ.3 ) THEN ITMP1 = NVAL( JI ) NVAL( JI ) = NVAL( KFNEW ) NVAL( KFNEW ) = ITMP1 END IF END IF KFNEW = KFNEW + 1 END IF 110 CONTINUE KF = KFNEW * * Choose Midpoints * DO 120 JI = KF, KL C( JI ) = HALF*( AB( JI, 1 )+AB( JI, 2 ) ) 120 CONTINUE * * If no more intervals to refine, quit. * IF( KF.GT.KL ) $ GO TO 140 130 CONTINUE * * Converged * 140 CONTINUE INFO = MAX( KL+1-KF, 0 ) MOUT = KL * RETURN * * End of DLAEBZ * END