My two cents on this. Like Matt, I make generous use of "always on" instruments. Cecilia is essentilly built around them and it does lead occasionally to obscure coding pratices. It would solve a number of parsing difficulties should the always running instruments be part of the header as opposed to to being instantiated like all others. But, on the verge on releasing Cecilia 3.0, I am not certain i want to completely revise my code base! The flag solution would work for now.

On another matter, I  would like to raise some flags with regards to cs5 stability. I am not quite certain yet if this is not due to my own TCL layer but I do get a LOT more seg faults in cs5. They seem to pop up out of the blue, ie, a module that normally works perfectly suddenly generates seg faults for a while and goes back to working ok without any changes being made. Do others find the same? If this turns out to be generalised, it may be an app killer. Should formal torture testing be envisaged?


Best to all

j

__________________________________________
http://jeanpiche.com

This might be a Tclcsound issue; if seg faults are occurring during 
performance,
eg. while a slider is being moved, or in general whenever a invalue set 
command
or a realtime score event is being sent to csound, *during* (ie. not 
before) performance.

As far as I know cecilia does not send score events (using csNote, csEvent, 
etc), am
I right? If so, I would suspect that if  the invalue channels are used and a
crash occurs, a memory conflict of sorts (the two threads trying to access 
the same
variable at the same time) could be happening. But as far as I can see this 
would be
unlikely. However if that's when it happens, then I can try to fix it with 
locks. But again,
the crashes are very specific; only when csPlay or csPlayAll has been 
invoked and
csInValue (or csNote, csEvent, etc. is used). Otherwise the problem is 
elsewhere.


Victor

At 13:20 18/09/2006, you wrote:
>On another matter, I  would like to raise some flags with regards to cs5 
>stability. I am not quite certain yet if this is not due to my own TCL 
>layer but I do get a LOT more seg faults in cs5. They seem to pop up out 
>of the blue, ie, a module that normally works perfectly suddenly generates 
>seg faults for a while and goes back to working ok without any changes 
>being made. Do others find the same? If this turns out to be generalised, 
>it may be an app killer. Should formal torture testing be envisaged?

Victor Lazzarini
Music Technology Laboratory
Music Department
National University of Ireland, Maynooth 


-------------------------------------------------------------------------
Using Tomcat but need to do more? Need to support web services, security?
Get stuff done quickly with pre-integrated technology to make your job easier
Download IBM WebSphere Application Server v.1.0.1 based on Apache Geronimo
http://sel.as-us.falkag.net/sel?cmd=lnk&kid=120709&bid=263057&dat=121642
_______________________________________________
Csound-devel mailing list
Csound-devel@lists.sourceforge.net

I have implemented mutex thread locks to avoid problems
in tclcsound multithreading. I did a basic test and they don't
seem to affect performance or cause any problems.

It would be worth trying to see if: 1) the changes
have not broken anything   and did not have an impact
on performance. 2) the changes solve the crashes
you were experiencing.

I am trying to commit but my network slow today. I
tried once and it timed out on me. As soon as I get it,
I'll let you know.

Victor

It 14:04 18/09/2006, you wrote:
>This might be a Tclcsound issue; if seg faults are occurring during
>performance,
>eg. while a slider is being moved, or in general whenever a invalue set
>command
>or a realtime score event is being sent to csound, *during* (ie. not
>before) performance.
>
>As far as I know cecilia does not send score events (using csNote, csEvent,
>etc), am
>I right? If so, I would suspect that if  the invalue channels are used and a
>crash occurs, a memory conflict of sorts (the two threads trying to access
>the same
>variable at the same time) could be happening. But as far as I can see this
>would be
>unlikely. However if that's when it happens, then I can try to fix it with
>locks. But again,
>the crashes are very specific; only when csPlay or csPlayAll has been
>invoked and
>csInValue (or csNote, csEvent, etc. is used). Otherwise the problem is
>elsewhere.
>
>
>Victor
>
>At 13:20 18/09/2006, you wrote:
> >On another matter, I  would like to raise some flags with regards to cs5
> >stability. I am not quite certain yet if this is not due to my own TCL
> >layer but I do get a LOT more seg faults in cs5. They seem to pop up out
> >of the blue, ie, a module that normally works perfectly suddenly generates
> >seg faults for a while and goes back to working ok without any changes
> >being made. Do others find the same? If this turns out to be generalised,
> >it may be an app killer. Should formal torture testing be envisaged?
>
>Victor Lazzarini
>Music Technology Laboratory
>Music Department
>National University of Ireland, Maynooth
>
>
>-------------------------------------------------------------------------
>Using Tomcat but need to do more? Need to support web services, security?
>Get stuff done quickly with pre-integrated technology to make your job easier
>Download IBM WebSphere Application Server v.1.0.1 based on Apache Geronimo
>http://sel.as-us.falkag.net/sel?cmd=lnk&kid=120709&bid=263057&dat=121642
>_______________________________________________
>Csound-devel mailing list
>Csound-devel@lists.sourceforge.net
>https://lists.sourceforge.net/lists/listinfo/csound-devel

Victor Lazzarini
Music Technology Laboratory
Music Department
National University of Ireland, Maynooth  


-------------------------------------------------------------------------
Using Tomcat but need to do more? Need to support web services, security?
Get stuff done quickly with pre-integrated technology to make your job easier
Download IBM WebSphere Application Server v.1.0.1 based on Apache Geronimo
http://sel.as-us.falkag.net/sel?cmd=lnk&kid=120709&bid=263057&dat=121642
_______________________________________________
Csound-devel mailing list
Csound-devel@lists.sourceforge.net

I have not had cs5 seg fault on me at all except when I am writing
duff code.  Can you provide some contexts?
==John ffitch

-------------------------------------------------------------------------
Take Surveys. Earn Cash. Influence the Future of IT
Join SourceForge.net's Techsay panel and you'll get the chance to share your
opinions on IT & business topics through brief surveys -- and earn cash
http://www.techsay.com/default.php?page=join.php&p=sourceforge&CID=DEVDEV
_______________________________________________
Csound-devel mailing list
Csound-devel@lists.sourceforge.net

I think this might be a threading issue in tclcsound. We are
working on it. If we find it's not, we'll come back to it.

At 09:54 20/09/2006, you wrote:
>I have not had cs5 seg fault on me at all except when I am writing
>duff code.  Can you provide some contexts?
>==John ffitch
>
>-------------------------------------------------------------------------
>Take Surveys. Earn Cash. Influence the Future of IT
>Join SourceForge.net's Techsay panel and you'll get the chance to share your
>opinions on IT & business topics through brief surveys -- and earn cash
>http://www.techsay.com/default.php?page=join.php&p=sourceforge&CID=DEVDEV
>_______________________________________________
>Csound-devel mailing list
>Csound-devel@lists.sourceforge.net
>https://lists.sourceforge.net/lists/listinfo/csound-devel

Victor Lazzarini
Music Technology Laboratory
Music Department
National University of Ireland, Maynooth 


-------------------------------------------------------------------------
Take Surveys. Earn Cash. Influence the Future of IT
Join SourceForge.net's Techsay panel and you'll get the chance to share your
opinions on IT & business topics through brief surveys -- and earn cash
http://www.techsay.com/default.php?page=join.php&p=sourceforge&CID=DEVDEV
_______________________________________________
Csound-devel mailing list
Csound-devel@lists.sourceforge.net