Re: sys2tf seems broken in 2.1.38

[John W. Eaton]

On  1-Nov-2002, Nicholas Piper <address@hidden> wrote:

| On Thu, 31 Oct 2002, John W. Eaton wrote:
| 
| > | error: A(I) = X: X must be a scalar or a vector with same length as I
| > | error: assignment failed, or no method for `matrix = matrix'
| 
| > Please try the following patch.  It should fix this problem as well as
| > another that caused other examples in DEMOcontrol to fail.
| 
| Thank you, that does appear to fix the DEMOcontrol (at least, that
| specific demo).
| 
| However, I'm still having troubles with the Control tools in general.
| My system is constructed from:
| 
| A=[0 1; -k/m -b/m];B=[0;1/m];C=[1 0];D=[0];
| sys = ss2sys(A,B,C,D)
| 
| sysout(sys,'tf') fails with :
|  ** On entry to DGESVD parameter number 13 had an illegal value
| Input(s)
|       1: u_1
| 
| Output(s):
|      1: y_1
| 
| error: exception encountered in Fortran subroutine dgesvd_
| error: unrecoverable error in dgesvd
| 
| [...]
| 
| Is this the right place to submit these errors, or should I use a bug
| mailing list ?

It would be better to send bug reports to the bug-octave mailing list.

| (It's possible I'm in error to try and call the
| functions I did, so maybe not :-) )

No, Octave should not be calling internal Lapack routines with invalid
parameters, so it is definitely a bug, but there's not enough
information above for me to reproduce the problem.  What are the
values of k and m?  I could guess, but I'd probably just guess wrong
and waste time.

jwe



-------------------------------------------------------------
Octave is freely available under the terms of the GNU GPL.

Octave's home on the web:  http://www.octave.org
How to fund new projects:  http://www.octave.org/funding.html
Subscription information:  http://www.octave.org/archive.html
-------------------------------------------------------------