[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: Changes in load from 2.1.50 to 2.1.57?
From: |
Miroslaw Kwasniak |
Subject: |
Re: Changes in load from 2.1.50 to 2.1.57? |
Date: |
Mon, 13 Dec 2004 10:11:50 +0100 |
User-agent: |
Mutt/1.5.6+20040907i |
On Mon, Dec 13, 2004 at 09:09:51AM +0100, Johansson Nicklas (RTLSN-NPJ) wrote:
> The engine toolbox in octave-forge contains a basic interface
> for calling Octave from within a C program. The interface uses
> a Unix FIFO to load data into Octave with the line:
>
> load -ascii -force "matfifo".
>
> where matfifo is the name of the FIFO.
>
> This works well with octave-2.1.50 on both Linux and Solaris,
> but fails for octave-2.1.57 and later versions in both OS.
I don't think this error is related to octave version - I can provoke it
even on octave 2.0 :(
It's not fully deterministic (failure rate can bee only few % or near 100%)
and it depend (???) on compiler / c++ library version and machine speed and
load. Simple - fifos and pipes arent full-prof in octave :(
One of many threads about this problem (not solved) you can see here:
http://www.octave.org/octave-lists/archive/help-octave.2004/msg02673.html
> Is it a bug or is there simply a better way to transport data
> from a C/C++ application to Octave, calculate and return the
> result back to C/C++ application?
Use files.
Mirek
-------------------------------------------------------------
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
-------------------------------------------------------------