DLACON(3)
π DLACON(3) β the reverse-communication 1-norm estimator β¨
okay so bestie, welcome to the Library Functions Manual corner of the group chat π. This page is for DLACON, and itβs giving βwe estimate a 1-norm together via reverse communicationβ main-character energy. Itβs not a solo mission β you and the subroutine keep passing X back and forth. Iconic. π¦
π NAME π
/home/abuild/rpmbuild/BUILD/lapack-static-3.12.0-build/lapack-3.12.0/SRC/dlacon.f
β¨ SYNOPSIS β¨
π§ Functions/Subroutines π§
subroutine DLACON (n, v, x, isgn, est, kase)β DLACON estimates the 1-norm of a square matrix, using reverse communication for evaluating matrix-vector products.
πΈ Function/Subroutine Documentation πΈ
Hereβs the signature, copied exactly because we do not mess with the source, periodt:
subroutine DLACON (integer n, double precision, dimension( * ) v, double precision, dimension( * ) x, integer, dimension( * ) isgn, double precision est, integer kase)
DLACON estimates the 1-norm of a square matrix, using reverse communication for evaluating matrix-vector products.
π Purpose π
The source literally says:
!>
!> DLACON estimates the 1-norm of a square, real matrix A.
!> Reverse communication is used for evaluating matrix-vector products.
!>
So yeah: square, real matrix A, 1-norm estimation, and reverse communication for matrix-vector products. Slay. π
π Parameters π
NβN is INTEGER; the order of the matrix.N >= 1. πVβV is DOUBLE PRECISION array, dimension (N). On the final return:
V = A*W, where EST = norm(V)/norm(W)
(W is not returned).
so W is lowkey not invited to the return party, but the formula is still iconic.
XβX is DOUBLE PRECISION array, dimension (N). On an intermediate return,Xshould be overwritten by:
A * X, if KASE=1,
A**T * X, if KASE=2,
and DLACON must be re-called with all the other parameters unchanged. No cap β same parameters, bestie. π«Ά
ISGNβISGN is INTEGER array, dimension (N). Thatβs the whole note; itβs giving mysterious side character. πESTβEST is DOUBLE PRECISION. On entry withKASE = 1or2andJUMP = 3,ESTshould be unchanged from the previous call toDLACON. On exit,ESTis an estimate (a lower bound) fornorm(A). Yes,JUMPappears in the note even though itβs not in the arg list β thatβs literally what the page says. πKASEβKASE is INTEGER. On the initial call toDLACON,KASEshould be0. On an intermediate return,KASEwill be1or2, indicating whetherXshould be overwritten byA * XorA**T * X. On the final return fromDLACON,KASEwill again be0. Full-circle moment. β
π Author π
- Univ. of Tennessee
- Univ. of California Berkeley
- Univ. of Colorado Denver
- NAG Ltd.
π«Ά Contributors π«Ά
- Nick Higham, University of Manchester.
- Originally named SONEST, dated March 16, 1988.
π References π
N.J. Higham, "FORTRAN codes for estimating the one-norm ofa real or complex matrix, with applications to condition estimation", ACM Trans. Math. Soft., vol. 14, no. 4, pp. 381-396, December 1988.
(That ofa is exactly how it appears, bestie β we preserve the sourceβs quirks. π
)
π§ Definition at line 114 of file dlacon.f π§
Definition at line 114 of file dlacon.f.
π€ Author (the Doxygen one) π€
Generated automatically by Doxygen for LAPACK from the source code.
π bye bestie! This page is LAPACK Version 3.12.0, source /home/abui...RC/dlacon.f(3). Stay slay, and remember: reverse communication is just a conversation with extra steps. β¨
π¦ where this page comes from
- opensuse-leap 16.0: lapack-man 3.12.0-160000.2.2 Β· BSD-3-Clause
- opensuse-leap 16.0: lapack-man 3.12.0-160000.2.2 Β· BSD-3-Clause
- opensuse-leap 16.0: lapack-man 3.12.0-160000.2.2 Β· BSD-3-Clause
- opensuse-leap 16.0: lapack-man 3.12.0-160000.2.2 Β· BSD-3-Clause
- π girlified by deepseek-ai/DeepSeek-V4.1-Flash-1017 on : a language model's rewrite of the original, under its terms