[Etherlab-users] Random Matlab crashes with Stateflow

[Andreas Stewering-Bone]

Dear Amos,

The message subsystem can cause some problems with the Etherlab
runtime.
I do not know this system exactly, but maybe the compilation fails
caused by missing dependencies in hrt_main.c (from Etherlab).
The message subsystem is a preperation for a middleware system and I
expect special runtime requirements for it outside the generated code.

If you use an actual EtherLab installation, you do not need the Classic
call interface anymore.

Am Mittwoch, dem 31.08.2022 um 08:00 +0000 schrieb Merkel, Amos:
> Dear Andreas,
>  
> Thanks a lot for your input! You definitely put me on the right track
> – the crash did not occur with the grt Target, so I searched for
> differences and found we had the classic call interface activated by
> default in our templates. That is what caused the crashes – without
> both grt and etherlab work, with Classic Call Interface activated,
> grt fails with an error message and etherlab crashes all of Matlab.
>  
> When I was wrote about messages, I meant the Simulink messages
> (https://de.mathworks.com/help/simulink/ug/simulink-messages-overview.html
> ). We use them to implement communication between state machines in
> our Controls.
>  
> Best regards and thanks again,
> Amos
>  
>  
> --
> Amos Merkel, M.Sc.
> Gruppenleiter
> Digitalisierung 
> 
> ifas – Institut für fluidtechnische Antriebe und Systeme
> RWTH Aachen University
> Campus-Boulevard 30
> 52074 Aachen
> 
> Tel: +49 241 80-47737
> Fax: +49 241 80-647712
> 
> www.ifas.rwth-aachen.de |Twitter |Facebook | YouTube |LinkedIn
>  
>  
>  
>  
>  
>  
> Von: Etherlab-users <etherlab-users-bounces at etherlab.org>Im Auftrag
> von Andreas Stewering-Bone
> Gesendet: Mittwoch, 31. August 2022 08:36
> An: etherlab-users at etherlab.org
> Betreff: Re: [Etherlab-users] Random Matlab crashes with Stateflow
>  
> Dear Amos,
>  
> The matlab library dependencies under linux are very specific and
> critical.
> You talked about an system upgrade, this can cause the crash because
> of an library incompatibility.
>  
> Etherlab itself has nearly no influence on the code generation and
> compilation. 
> A failure in Etherlab can only cause a compilation failure or a
> runtime failure in the included S-Functions.
>  
> Try to generate the code for the grt-Target. I think you will get the
> same result.
>  
> You talk about messages in a stateflow chart, Etherlab messages??
>  
> Best regards
>  
> Andreas
>  
> Am Dienstag, dem 30.08.2022 um 15:35 +0000 schrieb Merkel, Amos:
> > Hi,
> >  
> > correction: While the problem described below exists, the current
> > crashes are not random but related to the use of messages in
> > “Chart” stateflow blocks in Simulink.
> >  
> > The whole thing seems like a bug, I just cannot say if it is Matlab
> > or Etherlab. The problem didn’t exist a week or two ago, now a
> > Chart containing a message crashes Matlab when compiling with the
> > etherlab target. As I didn’t change my Matlab installation, but
> > pulled an update for my system some time ago, I guess the problem
> > is caused by some change in the etherlab toolchain.
> >  
> > If you are interested, you will find two minimum examples attached,
> > one that crashes and one that doesn’t crash. Note that once Matlab
> > crashed, it will crash every time until all compile files are
> > deleted (folders slprj and <model>_etl_hrt). I am using openSUSE
> > 15.2 with Linux 5.3.18-lp152.3.14-rt, the current etherlab
> > environment from the openSUSE repository and Matlab R2020a.
> >  
> > All the best,
> > Amos
> >  
> > --
> > Amos Merkel, M.Sc.
> > Gruppenleiter
> > Digitalisierung 
> > 
> > ifas – Institut für fluidtechnische Antriebe und Systeme
> > RWTH Aachen University
> > Campus-Boulevard 30
> > 52074 Aachen
> > 
> > Tel: +49 241 80-47737
> > Fax: +49 241 80-647712
> > 
> > www.ifas.rwth-aachen.de |Twitter |Facebook | YouTube |LinkedIn
> >  
> >  
> >  
> >  
> >  
> >  
> > Von: Etherlab-users <etherlab-users-bounces at etherlab.org>Im Auftrag
> > vonMerkel, Amos
> > Gesendet: Dienstag, 30. August 2022 13:41
> > An: etherlab-users at etherlab.org
> > Betreff: [Etherlab-users] Random Matlab crashes with Stateflow
> >  
> > Hi everyone,
> >  
> > I am using the Etherlab environment together with the etherlab
> > Simulink library. For the actual program procedure, I rely mostly
> > on the Stateflow blocks in Simulink. Usually  this works quite
> > well, but I have encountered random crashes while compiling several
> > times now. Typically it is an working Simulink file, you do a small
> > change and encounter the crashes. Reverting the change does not
> > solve the issue, however deleting the stateflow block does. Copy &
> > pasting the stateflow block to another model crashes that one as
> > well, reprogramming the stateflow block exactly the same, but from
> > scratch without copy+pasting anything works without problem.
> >  
> > Does anybody now this effect, or a better solution than programming
> > everything anew?
> >  
> > All the best,
> > Amos
> >  
> > --
> > Amos Merkel, M.Sc.
> > Group leader
> > Digitalization 
> > 
> > ifas – Institute for Fluid Power Drives and Systems
> > RWTH Aachen University
> > Campus-Boulevard 30
> > 52074 Aachen
> > 
> > Tel: +49 241 80-47737
> > Fax: +49 241 80-647712
> >  
> > www.ifas.rwth-aachen.de | Twitter | Facebook | YouTube | LinkedIn
> >  
> >  
> >  
> >  
> >  
> >  
> >  
> >  
> > --
> > Amos Merkel, M.Sc.
> > Gruppenleiter
> > Digitalisierung 
> > 
> > ifas – Institut für fluidtechnische Antriebe und Systeme
> > RWTH Aachen University
> > Campus-Boulevard 30
> > 52074 Aachen
> > 
> > Tel: +49 241 80-47737
> > Fax: +49 241 80-647712
> > 
> > www.ifas.rwth-aachen.de |Twitter |Facebook | YouTube |LinkedIn
>  

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.etherlab.org/pipermail/etherlab-users/attachments/20220831/17f149e2/attachment-0001.htm>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 659 bytes
Desc: This is a digitally signed message part
URL: <http://lists.etherlab.org/pipermail/etherlab-users/attachments/20220831/17f149e2/attachment-0001.sig>