Developer Reference for Intel® oneAPI Math Kernel Library for Fortran

ID 766686
Date 11/07/2023
Public

A newer version of this document is available. Customers should click here to go to the newest version.

Document Table of Contents

?laqr4

Computes the eigenvalues of a Hessenberg matrix, and optionally the matrices from the Schur decomposition.

Syntax

call slaqr4( wantt, wantz, n, ilo, ihi, h, ldh, wr, wi, iloz, ihiz, z, ldz, work, lwork, info )

call dlaqr4( wantt, wantz, n, ilo, ihi, h, ldh, wr, wi, iloz, ihiz, z, ldz, work, lwork, info )

call claqr4( wantt, wantz, n, ilo, ihi, h, ldh, w, iloz, ihiz, z, ldz, work, lwork, info )

call zlaqr4( wantt, wantz, n, ilo, ihi, h, ldh, w, iloz, ihiz, z, ldz, work, lwork, info )

Include Files

  • mkl.fi

Description

The routine computes the eigenvalues of a Hessenberg matrix H, and, optionally, the matrices T and Z from the Schur decomposition H=Z*T*ZH, where T is an upper quasi-triangular/triangular matrix (the Schur form), and Z is the orthogonal/unitary matrix of Schur vectors.

Optionally Z may be postmultiplied into an input orthogonal/unitary matrix Q so that this routine can give the Schur factorization of a matrix A which has been reduced to the Hessenberg form H by the orthogonal/unitary matrix Q: A = Q*H*QH = (QZ)*H*(QZ)H.

This routine implements one level of recursion for ?laqr0. It is a complete implementation of the small bulge multi-shift QR algorithm. It may be called by ?laqr0 and, for large enough deflation window size, it may be called by ?laqr3. This routine is identical to ?laqr0 except that it calls ?laqr2 instead of ?laqr3.

Input Parameters

wantt

LOGICAL.

If wantt = .TRUE., the full Schur form T is required;

If wantt = .FALSE., only eigenvalues are required.

wantz

LOGICAL.

If wantz = .TRUE., the matrix of Schur vectors Z is required;

If wantz = .FALSE., Schur vectors are not required.

n

INTEGER. The order of the Hessenberg matrix H. (n ≥ 0).

ilo, ihi

INTEGER.

It is assumed that H is already upper triangular in rows and columns 1:ilo-1 and ihi+1:n, and if ilo > 1 then h(ilo, ilo-1) = 0.

ilo and ihi are normally set by a previous call to cgebal, and then passed to cgehrd when the matrix output by cgebal is reduced to Hessenberg form. Otherwise, ilo and ihi should be set to 1 and n, respectively.

If n > 0, then 1 ≤ ilo ihi n.

If n=0, then ilo=1 and ihi=0

h

REAL for slaqr4

DOUBLE PRECISION for dlaqr4

COMPLEX for claqr4

DOUBLE COMPLEX for zlaqr4.

Array, DIMENSION (ldh, n), contains the upper Hessenberg matrix H.

ldh

INTEGER. The leading dimension of the array h. ldh ≥ max(1, n).

iloz, ihiz

INTEGER. Specify the rows of Z to which transformations must be applied if wantz is .TRUE., 1 ≤ iloz ilo; ihi ihiz n.

z

REAL for slaqr4

DOUBLE PRECISION for dlaqr4

COMPLEX for claqr4

DOUBLE COMPLEX for zlaqr4.

Array, DIMENSION (ldz, ihi), contains the matrix Z if wantz is .TRUE.. If wantz is .FALSE., z is not referenced.

ldz

INTEGER. The leading dimension of the array z.

If wantz is .TRUE., then ldz ≥ max(1, ihiz). Otherwise, ldz ≥ 1.

work

REAL for slaqr4

DOUBLE PRECISION for dlaqr4

COMPLEX for claqr4

DOUBLE COMPLEX for zlaqr4.

Workspace array with dimension lwork.

lwork

INTEGER. The dimension of the array work.

lwork ≥ max(1,n) is sufficient, but for the optimal performance a greater workspace may be required, typically as large as 6*n.

It is recommended to use the workspace query to determine the optimal workspace size. If lwork=-1,then the routine performs a workspace query: it estimates the optimal workspace size for the given values of the input parameters n, ilo, and ihi. The estimate is returned in work(1). No error messages related to the lwork is issued by xerbla. Neither H nor Z are accessed.

Output Parameters

h

If info=0, and wantt is .TRUE., then h contains the upper quasi-triangular/triangular matrix T from the Schur decomposition (the Schur form).

If info=0, and wantt is .FALSE., then the contents of h are unspecified on exit.

(The output values of h when info > 0 are given under the description of the info parameter below.)

The routines may explicitly set h(i,j) for i>j and j=1,2,...ilo-1 or j=ihi+1, ihi+2,...n.

work(1)

On exit work(1) contains the minimum value of lwork required for optimum performance.

w

COMPLEX for claqr4

DOUBLE COMPLEX for zlaqr4.

Arrays, DIMENSION(n). The computed eigenvalues of h(ilo:ihi, ilo:ihi) are stored in w(ilo:ihi). If wantt is .TRUE., then the eigenvalues are stored in the same order as on the diagonal of the Schur form returned in h, with w(i) = h(i,i).

wr, wi

REAL for slaqr4

DOUBLE PRECISION for dlaqr4

Arrays, DIMENSION(ihi) each. The real and imaginary parts, respectively, of the computed eigenvalues of h(ilo:ihi, ilo:ihi) are stored in the wr(ilo:ihi) and wi(ilo:ihi). If two eigenvalues are computed as a complex conjugate pair, they are stored in consecutive elements of wr and wi, say the i-th and (i+1)-th, with wi(i)> 0 and wi(i+1) < 0. If wantt is .TRUE., then the eigenvalues are stored in the same order as on the diagonal of the Schur form returned in h, with wr(i) = h(i,i), and if h(i:i+1,i:i+1)is a 2-by-2 diagonal block, then wi(i)=sqrt(-h(i+1,i)*h(i,i+1)).

z

If wantz is .TRUE., then z(ilo:ihi, iloz:ihiz) is replaced by z(ilo:ihi, iloz:ihiz)*U, where U is the orthogonal/unitary Schur factor of h(ilo:ihi, ilo:ihi).

If wantz is .FALSE., z is not referenced.

(The output values of z when info > 0 are given under the description of the info parameter below.)

info

INTEGER.

= 0: the execution is successful.

> 0: if info = i, then the routine failed to compute all the eigenvalues. Elements 1:ilo-1 and i+1:n of wr and wi contain those eigenvalues which have been successfully computed.

> 0: if wantt is .FALSE., then the remaining unconverged eigenvalues are the eigenvalues of the upper Hessenberg matrix rows and columns ilo through info of the final output value of h.

> 0: if wantt is .TRUE., then (initial value of h)*U = U*(final value of h, where U is an orthogonal/unitary matrix. The final value of h is upper Hessenberg and quasi-triangular/triangular in rows and columns info+1 through ihi.

> 0: if wantz is .TRUE., then (final value of z(ilo:ihi, iloz:ihiz))=(initial value of z(ilo:ihi, iloz:ihiz)*U, where U is the orthogonal/unitary matrix in the previous expression (regardless of the value of wantt).

> 0: if wantz is .FALSE., then z is not accessed.