# NAG Library Function Document

## 1Purpose

nag_dtgsyl (f08yhc) solves the generalized real quasi-triangular Sylvester equations.

## 2Specification

 #include #include
 void nag_dtgsyl (Nag_OrderType order, Nag_TransType trans, Integer ijob, Integer m, Integer n, const double a[], Integer pda, const double b[], Integer pdb, double c[], Integer pdc, const double d[], Integer pdd, const double e[], Integer pde, double f[], Integer pdf, double *scale, double *dif, NagError *fail)

## 3Description

nag_dtgsyl (f08yhc) solves either the generalized real Sylvester equations
 $AR-LB =αC DR-LE =αF,$ (1)
or the equations
 $ATR+DTL =αC RBT+LET =-αF,$ (2)
where the pair $\left(A,D\right)$ are given $m$ by $m$ matrices in real generalized Schur form, $\left(B,E\right)$ are given $n$ by $n$ matrices in real generalized Schur form and $\left(C,F\right)$ are given $m$ by $n$ matrices. The pair $\left(R,L\right)$ are the $m$ by $n$ solution matrices, and $\alpha$ is an output scaling factor determined by the function to avoid overflow in computing $\left(R,L\right)$.
Equations (1) are equivalent to equations of the form
 $Zx=αb ,$
where
 $Z = I⊗A-BT⊗I I⊗D-ET⊗I$
and $\otimes$ is the Kronecker product. Equations (2) are then equivalent to
 $ZTy = αb .$
The pair $\left(S,T\right)$ are in real generalized Schur form if $S$ is block upper triangular with $1$ by $1$ and $2$ by $2$ diagonal blocks on the diagonal and $T$ is upper triangular as returned, for example, by nag_dgges (f08xac), or nag_dhgeqz (f08xec) with ${\mathbf{job}}=\mathrm{Nag_Schur}$.
Optionally, the function estimates $\mathrm{Dif}\left[\left(A,D\right),\left(B,E\right)\right]$, the separation between the matrix pairs $\left(A,D\right)$ and $\left(B,E\right)$, which is the smallest singular value of $Z$. The estimate can be based on either the Frobenius norm, or the $1$-norm. The $1$-norm estimate can be three to ten times more expensive than the Frobenius norm estimate, but makes the condition estimation uniform with the nonsymmetric eigenproblem. The Frobenius norm estimate provides a low cost, but equally reliable estimate. For more information see Sections 2.4.8.3 and 4.11.1.3 of Anderson et al. (1999) and Kågström and Poromaa (1996).

## 4References

Anderson E, Bai Z, Bischof C, Blackford S, Demmel J, Dongarra J J, Du Croz J J, Greenbaum A, Hammarling S, McKenney A and Sorensen D (1999) LAPACK Users' Guide (3rd Edition) SIAM, Philadelphia http://www.netlib.org/lapack/lug
Kågström B (1994) A perturbation analysis of the generalized Sylvester equation $\left(AR-LB,DR-LE\right)=\left(c,F\right)$ SIAM J. Matrix Anal. Appl. 15 1045–1060
Kågström B and Poromaa P (1996) LAPACK-style algorithms and software for solving the generalized Sylvester equation and estimating the separation between regular matrix pairs ACM Trans. Math. Software 22 78–103

## 5Arguments

1:    $\mathbf{order}$Nag_OrderTypeInput
On entry: the order argument specifies the two-dimensional storage scheme being used, i.e., row-major ordering or column-major ordering. C language defined storage is specified by ${\mathbf{order}}=\mathrm{Nag_RowMajor}$. See Section 3.3.1.3 in How to Use the NAG Library and its Documentation for a more detailed explanation of the use of this argument.
Constraint: ${\mathbf{order}}=\mathrm{Nag_RowMajor}$ or $\mathrm{Nag_ColMajor}$.
2:    $\mathbf{trans}$Nag_TransTypeInput
On entry: if ${\mathbf{trans}}=\mathrm{Nag_NoTrans}$, solve the generalized Sylvester equation (1).
If ${\mathbf{trans}}=\mathrm{Nag_Trans}$, solve the ‘transposed’ system (2).
Constraint: ${\mathbf{trans}}=\mathrm{Nag_NoTrans}$ or $\mathrm{Nag_Trans}$.
3:    $\mathbf{ijob}$IntegerInput
On entry: specifies what kind of functionality is to be performed when ${\mathbf{trans}}=\mathrm{Nag_NoTrans}$.
${\mathbf{ijob}}=0$
Solve (1) only.
${\mathbf{ijob}}=1$
The functionality of ${\mathbf{ijob}}=0$ and $3$.
${\mathbf{ijob}}=2$
The functionality of ${\mathbf{ijob}}=0$ and $4$.
${\mathbf{ijob}}=3$
Only an estimate of $\mathrm{Dif}\left[\left(A,D\right),\left(B,E\right)\right]$ is computed based on the Frobenius norm.
${\mathbf{ijob}}=4$
Only an estimate of $\mathrm{Dif}\left[\left(A,D\right),\left(B,E\right)\right]$ is computed based on the $1$-norm.
If ${\mathbf{trans}}=\mathrm{Nag_Trans}$, ijob is not referenced.
Constraint: if ${\mathbf{trans}}=\mathrm{Nag_NoTrans}$, $0\le {\mathbf{ijob}}\le 4$.
4:    $\mathbf{m}$IntegerInput
On entry: $m$, the order of the matrices $A$ and $D$, and the row dimension of the matrices $C$, $F$, $R$ and $L$.
Constraint: ${\mathbf{m}}>0$.
5:    $\mathbf{n}$IntegerInput
On entry: $n$, the order of the matrices $B$ and $E$, and the column dimension of the matrices $C$, $F$, $R$ and $L$.
Constraint: ${\mathbf{n}}>0$.
6:    $\mathbf{a}\left[\mathit{dim}\right]$const doubleInput
Note: the dimension, dim, of the array a must be at least $\mathrm{max}\phantom{\rule{0.125em}{0ex}}\left(1,{\mathbf{pda}}×{\mathbf{m}}\right)$.
The $\left(i,j\right)$th element of the matrix $A$ is stored in
• ${\mathbf{a}}\left[\left(j-1\right)×{\mathbf{pda}}+i-1\right]$ when ${\mathbf{order}}=\mathrm{Nag_ColMajor}$;
• ${\mathbf{a}}\left[\left(i-1\right)×{\mathbf{pda}}+j-1\right]$ when ${\mathbf{order}}=\mathrm{Nag_RowMajor}$.
On entry: the upper quasi-triangular matrix $A$.
7:    $\mathbf{pda}$IntegerInput
On entry: the stride separating row or column elements (depending on the value of order) in the array a.
Constraint: ${\mathbf{pda}}\ge \mathrm{max}\phantom{\rule{0.125em}{0ex}}\left(1,{\mathbf{m}}\right)$.
8:    $\mathbf{b}\left[\mathit{dim}\right]$const doubleInput
Note: the dimension, dim, of the array b must be at least $\mathrm{max}\phantom{\rule{0.125em}{0ex}}\left(1,{\mathbf{pdb}}×{\mathbf{n}}\right)$.
The $\left(i,j\right)$th element of the matrix $B$ is stored in
• ${\mathbf{b}}\left[\left(j-1\right)×{\mathbf{pdb}}+i-1\right]$ when ${\mathbf{order}}=\mathrm{Nag_ColMajor}$;
• ${\mathbf{b}}\left[\left(i-1\right)×{\mathbf{pdb}}+j-1\right]$ when ${\mathbf{order}}=\mathrm{Nag_RowMajor}$.
On entry: the upper quasi-triangular matrix $B$.
9:    $\mathbf{pdb}$IntegerInput
On entry: the stride separating row or column elements (depending on the value of order) in the array b.
Constraint: ${\mathbf{pdb}}\ge \mathrm{max}\phantom{\rule{0.125em}{0ex}}\left(1,{\mathbf{n}}\right)$.
10:  $\mathbf{c}\left[\mathit{dim}\right]$doubleInput/Output
Note: the dimension, dim, of the array c must be at least
• $\mathrm{max}\phantom{\rule{0.125em}{0ex}}\left(1,{\mathbf{pdc}}×{\mathbf{n}}\right)$ when ${\mathbf{order}}=\mathrm{Nag_ColMajor}$;
• $\mathrm{max}\phantom{\rule{0.125em}{0ex}}\left(1,{\mathbf{m}}×{\mathbf{pdc}}\right)$ when ${\mathbf{order}}=\mathrm{Nag_RowMajor}$.
The $\left(i,j\right)$th element of the matrix $C$ is stored in
• ${\mathbf{c}}\left[\left(j-1\right)×{\mathbf{pdc}}+i-1\right]$ when ${\mathbf{order}}=\mathrm{Nag_ColMajor}$;
• ${\mathbf{c}}\left[\left(i-1\right)×{\mathbf{pdc}}+j-1\right]$ when ${\mathbf{order}}=\mathrm{Nag_RowMajor}$.
On entry: contains the right-hand-side matrix $C$.
On exit: if ${\mathbf{ijob}}=0$, $1$ or $2$, c is overwritten by the solution matrix $R$.
If ${\mathbf{trans}}=\mathrm{Nag_NoTrans}$ and ${\mathbf{ijob}}=3$ or $4$, c holds $R$, the solution achieved during the computation of the Dif estimate.
11:  $\mathbf{pdc}$IntegerInput
On entry: the stride separating row or column elements (depending on the value of order) in the array c.
Constraints:
• if ${\mathbf{order}}=\mathrm{Nag_ColMajor}$, ${\mathbf{pdc}}\ge \mathrm{max}\phantom{\rule{0.125em}{0ex}}\left(1,{\mathbf{m}}\right)$;
• if ${\mathbf{order}}=\mathrm{Nag_RowMajor}$, ${\mathbf{pdc}}\ge \mathrm{max}\phantom{\rule{0.125em}{0ex}}\left(1,{\mathbf{n}}\right)$.
12:  $\mathbf{d}\left[\mathit{dim}\right]$const doubleInput
Note: the dimension, dim, of the array d must be at least $\mathrm{max}\phantom{\rule{0.125em}{0ex}}\left(1,{\mathbf{pdd}}×{\mathbf{m}}\right)$.
The $\left(i,j\right)$th element of the matrix $D$ is stored in
• ${\mathbf{d}}\left[\left(j-1\right)×{\mathbf{pdd}}+i-1\right]$ when ${\mathbf{order}}=\mathrm{Nag_ColMajor}$;
• ${\mathbf{d}}\left[\left(i-1\right)×{\mathbf{pdd}}+j-1\right]$ when ${\mathbf{order}}=\mathrm{Nag_RowMajor}$.
On entry: the upper triangular matrix $D$.
13:  $\mathbf{pdd}$IntegerInput
On entry: the stride separating row or column elements (depending on the value of order) in the array d.
Constraint: ${\mathbf{pdd}}\ge \mathrm{max}\phantom{\rule{0.125em}{0ex}}\left(1,{\mathbf{m}}\right)$.
14:  $\mathbf{e}\left[\mathit{dim}\right]$const doubleInput
Note: the dimension, dim, of the array e must be at least $\mathrm{max}\phantom{\rule{0.125em}{0ex}}\left(1,{\mathbf{pde}}×{\mathbf{n}}\right)$.
The $\left(i,j\right)$th element of the matrix $E$ is stored in
• ${\mathbf{e}}\left[\left(j-1\right)×{\mathbf{pde}}+i-1\right]$ when ${\mathbf{order}}=\mathrm{Nag_ColMajor}$;
• ${\mathbf{e}}\left[\left(i-1\right)×{\mathbf{pde}}+j-1\right]$ when ${\mathbf{order}}=\mathrm{Nag_RowMajor}$.
On entry: the upper triangular matrix $E$.
15:  $\mathbf{pde}$IntegerInput
On entry: the stride separating row or column elements (depending on the value of order) in the array e.
Constraint: ${\mathbf{pde}}\ge \mathrm{max}\phantom{\rule{0.125em}{0ex}}\left(1,{\mathbf{n}}\right)$.
16:  $\mathbf{f}\left[\mathit{dim}\right]$doubleInput/Output
Note: the dimension, dim, of the array f must be at least
• $\mathrm{max}\phantom{\rule{0.125em}{0ex}}\left(1,{\mathbf{pdf}}×{\mathbf{n}}\right)$ when ${\mathbf{order}}=\mathrm{Nag_ColMajor}$;
• $\mathrm{max}\phantom{\rule{0.125em}{0ex}}\left(1,{\mathbf{m}}×{\mathbf{pdf}}\right)$ when ${\mathbf{order}}=\mathrm{Nag_RowMajor}$.
The $\left(i,j\right)$th element of the matrix $F$ is stored in
• ${\mathbf{f}}\left[\left(j-1\right)×{\mathbf{pdf}}+i-1\right]$ when ${\mathbf{order}}=\mathrm{Nag_ColMajor}$;
• ${\mathbf{f}}\left[\left(i-1\right)×{\mathbf{pdf}}+j-1\right]$ when ${\mathbf{order}}=\mathrm{Nag_RowMajor}$.
On entry: contains the right-hand side matrix $F$.
On exit: if ${\mathbf{ijob}}=0$, $1$ or $2$, f is overwritten by the solution matrix $L$.
If ${\mathbf{trans}}=\mathrm{Nag_NoTrans}$ and ${\mathbf{ijob}}=3$ or $4$, f holds $L$, the solution achieved during the computation of the Dif estimate.
17:  $\mathbf{pdf}$IntegerInput
On entry: the stride separating row or column elements (depending on the value of order) in the array f.
Constraints:
• if ${\mathbf{order}}=\mathrm{Nag_ColMajor}$, ${\mathbf{pdf}}\ge \mathrm{max}\phantom{\rule{0.125em}{0ex}}\left(1,{\mathbf{m}}\right)$;
• if ${\mathbf{order}}=\mathrm{Nag_RowMajor}$, ${\mathbf{pdf}}\ge \mathrm{max}\phantom{\rule{0.125em}{0ex}}\left(1,{\mathbf{n}}\right)$.
18:  $\mathbf{scale}$double *Output
On exit: $\alpha$, the scaling factor in (1) or (2).
If $0<{\mathbf{scale}}<1$, c and f hold the solutions $R$ and $L$, respectively, to a slightly perturbed system but the input arrays a, b, d and e have not been changed.
If ${\mathbf{scale}}=0$, c and f hold the solutions $R$ and $L$, respectively, to the homogeneous system with $C=F=0$. In this case dif is not referenced.
Normally, ${\mathbf{scale}}=1$.
19:  $\mathbf{dif}$double *Output
On exit: the estimate of $\mathrm{Dif}$. If ${\mathbf{ijob}}=0$, dif is not referenced.
20:  $\mathbf{fail}$NagError *Input/Output
The NAG error argument (see Section 3.7 in How to Use the NAG Library and its Documentation).

## 6Error Indicators and Warnings

NE_ALLOC_FAIL
Dynamic memory allocation failed.
See Section 2.3.1.2 in How to Use the NAG Library and its Documentation for further information.
On entry, argument $〈\mathit{\text{value}}〉$ had an illegal value.
NE_EIGENVALUES
$\left(A,D\right)$ and $\left(B,E\right)$ have common or close eigenvalues and so no solution could be computed.
NE_ENUM_INT
On entry, ${\mathbf{trans}}=〈\mathit{\text{value}}〉$ and ${\mathbf{ijob}}=〈\mathit{\text{value}}〉$.
Constraint: if ${\mathbf{trans}}=\mathrm{Nag_NoTrans}$, $0\le {\mathbf{ijob}}\le 4$.
NE_INT
On entry, ${\mathbf{m}}=〈\mathit{\text{value}}〉$.
Constraint: ${\mathbf{m}}>0$.
On entry, ${\mathbf{n}}=〈\mathit{\text{value}}〉$.
Constraint: ${\mathbf{n}}>0$.
On entry, ${\mathbf{pda}}=〈\mathit{\text{value}}〉$.
Constraint: ${\mathbf{pda}}>0$.
On entry, ${\mathbf{pdb}}=〈\mathit{\text{value}}〉$.
Constraint: ${\mathbf{pdb}}>0$.
On entry, ${\mathbf{pdc}}=〈\mathit{\text{value}}〉$.
Constraint: ${\mathbf{pdc}}>0$.
On entry, ${\mathbf{pdd}}=〈\mathit{\text{value}}〉$.
Constraint: ${\mathbf{pdd}}>0$.
On entry, ${\mathbf{pde}}=〈\mathit{\text{value}}〉$.
Constraint: ${\mathbf{pde}}>0$.
On entry, ${\mathbf{pdf}}=〈\mathit{\text{value}}〉$.
Constraint: ${\mathbf{pdf}}>0$.
NE_INT_2
On entry, ${\mathbf{pda}}=〈\mathit{\text{value}}〉$ and ${\mathbf{m}}=〈\mathit{\text{value}}〉$.
Constraint: ${\mathbf{pda}}\ge \mathrm{max}\phantom{\rule{0.125em}{0ex}}\left(1,{\mathbf{m}}\right)$.
On entry, ${\mathbf{pdb}}=〈\mathit{\text{value}}〉$ and ${\mathbf{n}}=〈\mathit{\text{value}}〉$.
Constraint: ${\mathbf{pdb}}\ge \mathrm{max}\phantom{\rule{0.125em}{0ex}}\left(1,{\mathbf{n}}\right)$.
On entry, ${\mathbf{pdc}}=〈\mathit{\text{value}}〉$ and ${\mathbf{m}}=〈\mathit{\text{value}}〉$.
Constraint: ${\mathbf{pdc}}\ge \mathrm{max}\phantom{\rule{0.125em}{0ex}}\left(1,{\mathbf{m}}\right)$.
On entry, ${\mathbf{pdc}}=〈\mathit{\text{value}}〉$ and ${\mathbf{n}}=〈\mathit{\text{value}}〉$.
Constraint: ${\mathbf{pdc}}\ge \mathrm{max}\phantom{\rule{0.125em}{0ex}}\left(1,{\mathbf{n}}\right)$.
On entry, ${\mathbf{pdd}}=〈\mathit{\text{value}}〉$ and ${\mathbf{m}}=〈\mathit{\text{value}}〉$.
Constraint: ${\mathbf{pdd}}\ge \mathrm{max}\phantom{\rule{0.125em}{0ex}}\left(1,{\mathbf{m}}\right)$.
On entry, ${\mathbf{pde}}=〈\mathit{\text{value}}〉$ and ${\mathbf{n}}=〈\mathit{\text{value}}〉$.
Constraint: ${\mathbf{pde}}\ge \mathrm{max}\phantom{\rule{0.125em}{0ex}}\left(1,{\mathbf{n}}\right)$.
On entry, ${\mathbf{pdf}}=〈\mathit{\text{value}}〉$ and ${\mathbf{m}}=〈\mathit{\text{value}}〉$.
Constraint: ${\mathbf{pdf}}\ge \mathrm{max}\phantom{\rule{0.125em}{0ex}}\left(1,{\mathbf{m}}\right)$.
On entry, ${\mathbf{pdf}}=〈\mathit{\text{value}}〉$ and ${\mathbf{n}}=〈\mathit{\text{value}}〉$.
Constraint: ${\mathbf{pdf}}\ge \mathrm{max}\phantom{\rule{0.125em}{0ex}}\left(1,{\mathbf{n}}\right)$.
NE_INTERNAL_ERROR
An internal error has occurred in this function. Check the function call and any array sizes. If the call is correct then please contact NAG for assistance.
See Section 2.7.6 in How to Use the NAG Library and its Documentation for further information.
NE_NO_LICENCE
Your licence key may have expired or may not have been installed correctly.
See Section 2.7.5 in How to Use the NAG Library and its Documentation for further information.

## 7Accuracy

See Kågström (1994) for a perturbation analysis of the generalized Sylvester equation.

## 8Parallelism and Performance

nag_dtgsyl (f08yhc) makes calls to BLAS and/or LAPACK routines, which may be threaded within the vendor library used by this implementation. Consult the documentation for the vendor library for further information.
Please consult the x06 Chapter Introduction for information on how to control and interrogate the OpenMP environment used within this function. Please also consult the Users' Note for your implementation for any additional implementation-specific information.

The total number of floating-point operations needed to solve the generalized Sylvester equations is approximately $2mn\left(n+m\right)$. The Frobenius norm estimate of $\mathrm{Dif}$ does not require additional significant computation, but the $1$-norm estimate is typically five times more expensive.
The complex analogue of this function is nag_ztgsyl (f08yvc).

## 10Example

This example solves the generalized Sylvester equations
 $AR-LB =αC DR-LE =αF,$
where
 $A = 4.0 1.0 1.0 2.0 0.0 3.0 4.0 1.0 0.0 1.0 3.0 1.0 0.0 0.0 0.0 6.0 , B= 1.0 1.0 1.0 1.0 0.0 3.0 4.0 1.0 0.0 1.0 3.0 1.0 0.0 0.0 0.0 4.0 ,$
 $D = 2.0 1.0 1.0 3.0 0.0 1.0 2.0 1.0 0.0 0.0 1.0 1.0 0.0 0.0 0.0 2.0 , E= 1.0 1.0 1.0 2.0 0.0 1.0 4.0 1.0 0.0 0.0 1.0 1.0 0.0 0.0 0.0 1.0 ,$
 $C = -4.0 7.0 1.0 12.0 -9.0 2.0 -2.0 -2.0 -4.0 2.0 -2.0 8.0 -7.0 7.0 -6.0 19.0 and F= -7.0 5.0 0.0 7.0 -5.0 1.0 -8.0 0.0 -1.0 2.0 -3.0 5.0 -3.0 2.0 0.0 5.0 .$

### 10.1Program Text

Program Text (f08yhce.c)

### 10.2Program Data

Program Data (f08yhce.d)

### 10.3Program Results

Program Results (f08yhce.r)

© The Numerical Algorithms Group Ltd, Oxford, UK. 2017