I am  getting quite desperate here. I have no idea why portmidi  
stopped working on OSX. The debugger seems to point to a  
pthread_start() function that I cannot locate. It's not in portmidi,  
but maybe in the CoreMIDI library. The error message I am getting from  
somewhere

2010-03-01 22:00:39.086 cstest[10980:10b] *** CFMessagePort:  
bootstrap_register(): failed 1103 (0x44f) 'Service name already  
exists', port = 0x5f1b, name = 'Apple MIDI Client'
See /usr/include/servers/bootstrap_defs.h for the error codes.

seems to be related to thread communication (CFMessagePort does that).

The strangest thing is that with CsoundLib 5.11 there is no problem,  
no error messages and everything just works. What changes in CsoundLib  
could have brought about this problem? Note that there is nothing in  
pmidi.c that could have caused that (the file has not changed).

Victor 

------------------------------------------------------------------------------
Download Intel® Parallel Studio Eval
Try the new software tools for yourself. Speed compiling, find bugs
proactively, and fine-tune applications for parallel performance.
See why Intel Parallel Studio got high marks during beta.
http://p.sf.net/sfu/intel-sw-dev
_______________________________________________
Csound-devel mailing list
Csound-devel@lists.sourceforge.net

You can do a bisect of CVS between 5.11 and 5.12 to try to find the
guilty commit.

Basically do this:

good_date = 5.11 release date
bad_date = 5.12 release date

do:
    pick a date between good_date and bad_date
    checkout CVS from that date, build and test it
    if it fails, then bad_date = current_date
    else,  good_date = current_date

repeat until you have no more commits left.

CVS commits each file separately, which is why you have to iterate
over dates instead of commits, because so far the precise file that
causes the failure is unknown.


http://wingolog.org/archives/2006/01/20/cvs-bisect seems to have a
tool to help automating this search.


On Mon, Mar 1, 2010 at 19:15, Victor Lazzarini  wrote:
> Note that there is nothing in
> pmidi.c that could have caused that (the file has not changed).



-- 

Saludos,
Felipe Sateler

------------------------------------------------------------------------------
Download Intel® Parallel Studio Eval
Try the new software tools for yourself. Speed compiling, find bugs
proactively, and fine-tune applications for parallel performance.
See why Intel Parallel Studio got high marks during beta.
http://p.sf.net/sfu/intel-sw-dev
_______________________________________________
Csound-devel mailing list
Csound-devel@lists.sourceforge.net

Thanks, Felipe. That did it.

In fact, it was something I suspected, because it was the only OSX- 
specific change that was vaguely related.
The problem is in Top/csmodule.c: the old MACH code for loading  
modules was changed (by me) to use the
more standard dlopen() interface (as in Linux).

However this seems to cause the PortMIDI module to fail under more  
than one instance. It does not seem to be an easy fix, because the old  
code uses lots of deprecated functions (that do not work in OSX10.6,  
for instance, so that is why I changed it).

So I'll have to think of a solution for this. Not an easy one, though,  
but maybe better than having to write a new MIDI module.

Regards

Victor



On 1 Mar 2010, at 22:24, Felipe Sateler wrote:

> You can do a bisect of CVS between 5.11 and 5.12 to try to find the
> guilty commit.
>
> Basically do this:
>
> good_date = 5.11 release date
> bad_date = 5.12 release date
>
> do:
>    pick a date between good_date and bad_date
>    checkout CVS from that date, build and test it
>    if it fails, then bad_date = current_date
>    else,  good_date = current_date
>
> repeat until you have no more commits left.
>
> CVS commits each file separately, which is why you have to iterate
> over dates instead of commits, because so far the precise file that
> causes the failure is unknown.
>
>
> http://wingolog.org/archives/2006/01/20/cvs-bisect seems to have a
> tool to help automating this search.
>
>
> On Mon, Mar 1, 2010 at 19:15, Victor Lazzarini  > wrote:
>> Note that there is nothing in
>> pmidi.c that could have caused that (the file has not changed).
>
>
>
> -- 
>
> Saludos,
> Felipe Sateler
>
> ------------------------------------------------------------------------------
> Download Intel® Parallel Studio Eval
> Try the new software tools for yourself. Speed compiling, find bugs
> proactively, and fine-tune applications for parallel performance.
> See why Intel Parallel Studio got high marks during beta.
> http://p.sf.net/sfu/intel-sw-dev
> _______________________________________________
> Csound-devel mailing list
> Csound-devel@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/csound-devel


------------------------------------------------------------------------------
Download Intel® Parallel Studio Eval
Try the new software tools for yourself. Speed compiling, find bugs
proactively, and fine-tune applications for parallel performance.
See why Intel Parallel Studio got high marks during beta.
http://p.sf.net/sfu/intel-sw-dev
_______________________________________________
Csound-devel mailing list
Csound-devel@lists.sourceforge.net

Hi Victor,

Your fix has got rid of the crashes, thanks! But it seems it's still
not fully reentrant, and a second instance will not run because it
can't open the MIDI device:
It will say:

The available MIDI in devices are:
   0: IAC Driver Bus 1 (CoreMIDI)
PortMIDI: Activated input device 0: 'IAC Driver Bus 1' (CoreMIDI)
 *** PortMIDI: error opening input device 0: PortMidi: `Bad pointer'
 *** error opening MIDI in device: -1 (Unknown MIDI error)

Anyway, it's a huge improvement.

Cheers,
Andrés

On Mon, Mar 1, 2010 at 11:37 PM, Victor Lazzarini
 wrote:
> Thanks, Felipe. That did it.
>
> In fact, it was something I suspected, because it was the only OSX-
> specific change that was vaguely related.
> The problem is in Top/csmodule.c: the old MACH code for loading
> modules was changed (by me) to use the
> more standard dlopen() interface (as in Linux).
>
> However this seems to cause the PortMIDI module to fail under more
> than one instance. It does not seem to be an easy fix, because the old
> code uses lots of deprecated functions (that do not work in OSX10.6,
> for instance, so that is why I changed it).
>
> So I'll have to think of a solution for this. Not an easy one, though,
> but maybe better than having to write a new MIDI module.
>
> Regards
>
> Victor
>
>
>
> On 1 Mar 2010, at 22:24, Felipe Sateler wrote:
>
>> You can do a bisect of CVS between 5.11 and 5.12 to try to find the
>> guilty commit.
>>
>> Basically do this:
>>
>> good_date = 5.11 release date
>> bad_date = 5.12 release date
>>
>> do:
>>    pick a date between good_date and bad_date
>>    checkout CVS from that date, build and test it
>>    if it fails, then bad_date = current_date
>>    else,  good_date = current_date
>>
>> repeat until you have no more commits left.
>>
>> CVS commits each file separately, which is why you have to iterate
>> over dates instead of commits, because so far the precise file that
>> causes the failure is unknown.
>>
>>
>> http://wingolog.org/archives/2006/01/20/cvs-bisect seems to have a
>> tool to help automating this search.
>>
>>
>> On Mon, Mar 1, 2010 at 19:15, Victor Lazzarini > > wrote:
>>> Note that there is nothing in
>>> pmidi.c that could have caused that (the file has not changed).
>>
>>
>>
>> --
>>
>> Saludos,
>> Felipe Sateler
>>
>> ------------------------------------------------------------------------------
>> Download Intel® Parallel Studio Eval
>> Try the new software tools for yourself. Speed compiling, find bugs
>> proactively, and fine-tune applications for parallel performance.
>> See why Intel Parallel Studio got high marks during beta.
>> http://p.sf.net/sfu/intel-sw-dev
>> _______________________________________________
>> Csound-devel mailing list
>> Csound-devel@lists.sourceforge.net
>> https://lists.sourceforge.net/lists/listinfo/csound-devel
>
>
> ------------------------------------------------------------------------------
> Download Intel® Parallel Studio Eval
> Try the new software tools for yourself. Speed compiling, find bugs
> proactively, and fine-tune applications for parallel performance.
> See why Intel Parallel Studio got high marks during beta.
> http://p.sf.net/sfu/intel-sw-dev
> _______________________________________________
> Csound-devel mailing list
> Csound-devel@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/csound-devel
>



-- 


Andrés

------------------------------------------------------------------------------
Download Intel® Parallel Studio Eval
Try the new software tools for yourself. Speed compiling, find bugs
proactively, and fine-tune applications for parallel performance.
See why Intel Parallel Studio got high marks during beta.
http://p.sf.net/sfu/intel-sw-dev
_______________________________________________
Csound-devel mailing list
Csound-devel@lists.sourceforge.net
https://lists.sourceforge.net/

Hi,

On Tue, Mar 2, 2010 at 6:50 PM,   wrote:
> yes, that seems right to me. The library code is not re-entrant, but should work in sequenced calls.
> I see that Istvan has some code to protect the Pm_Terminate() function against being called by an instance before all instances are closed, and a similar thing to Pm_Initialize(). However, the globals in the portmidi library would clash. Interesting that they don't on Linux (and Windows?).
>

I haven't checked whether portmidi is reentrant on linux, since I use
alsa, and that works fine. I'll check soon.

I hope you can find a way to make a reentrant portmidi module, as
things in QuteCsound are working well now, and you can run several
instances without a problem on OS X and Linux using jack, which is
great.

Cheers,
Andrés

------------------------------------------------------------------------------
Download Intel® Parallel Studio Eval
Try the new software tools for yourself. Speed compiling, find bugs
proactively, and fine-tune applications for parallel performance.
See why Intel Parallel Studio got high marks during beta.
http://p.sf.net/sfu/intel-sw-dev
_______________________________________________
Csound-devel mailing list
Csound-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/

2010/3/2  :
>...However, the globals in the portmidi library would clash. Interesting that they don't on Linux (and Windows?

Maybe related:
Several of the recent versions of Csound on Windows will not close the
portmidi midi device properly on exit. Michael Gogins knows more about
the details on this. The issues was reported to be fixed, then
reappeared. I've started using -+rtmidi=mme as the default, as this
avoids the problems. I don't know if it's related to the OSX portmidi
problems, but it's got to do with portmidi and application exit
anyway...

Oeyvind

------------------------------------------------------------------------------
Download Intel® Parallel Studio Eval
Try the new software tools for yourself. Speed compiling, find bugs
proactively, and fine-tune applications for parallel performance.
See why Intel Parallel Studio got high marks during beta.
http://p.sf.net/sfu/intel-sw-dev
_______________________________________________
Csound-devel mailing list
Csound-devel@lists.sourceforge.net

Hi,

I thought you said portaudio would have the same problems as coreaudio
if it used the coreaudio interface, so I disabled running multiple
instances with portaudio. Would wrapping coreaudio in pulseaudio make
coreaudio reentrant?

Or should coreaudio work already?

Cheers,
Andrés

On Tue, Mar 2, 2010 at 10:20 PM,   wrote:
> What about using portaudio? That should be reentrant too, with the new
> version of portaudio I'm distributing.
>
> Victor
>
> ----- Original Message -----
> From: Andres Cabrera 
> Date: Tuesday, March 2, 2010 8:17 pm
> Subject: Re: [Cs-dev] Found the b****r (was Re: portmidi, still no answers
> in sight)
> To: Developer discussions 
>
>> Hi,
>>
>> On Tue, Mar 2, 2010 at 6:50 PM,
>>  wrote:
>> > yes, that seems right to me. The library code is not re-
>> entrant, but should work in sequenced calls.
>> > I see that Istvan has some code to protect the Pm_Terminate()
>> function against being called by an instance before all
>> instances are closed, and a similar thing to Pm_Initialize().
>> However, the globals in the portmidi library would clash.
>> Interesting that they don't on Linux (and Windows?).
>> >
>>
>> I haven't checked whether portmidi is reentrant on linux, since
>> I use
>> alsa, and that works fine. I'll check soon.
>>
>> I hope you can find a way to make a reentrant portmidi module, as
>> things in QuteCsound are working well now, and you can run several
>> instances without a problem on OS X and Linux using jack, which is
>> great.
>>
>> Cheers,
>> Andrés
>>
>> -----------------------------------------------------------------
>> -------------
>> Download Intel® Parallel Studio Eval
>> Try the new software tools for yourself. Speed compiling, find bugs
>> proactively, and fine-tune applications for parallel performance.
>> See why Intel Parallel Studio got high marks during beta.
>> http://p.sf.net/sfu/intel-sw-dev
>> _______________________________________________
>> Csound-devel mailing list
>> Csound-devel@lists.sourceforge.net
>> https://lists.sourceforge.net/lists/listinfo/csound-devel
>
> Dr Victor Lazzarini, Senior Lecturer, Dept. of Music,
> National University of Ireland, Maynooth
>
> ------------------------------------------------------------------------------
> Download Intel® Parallel Studio Eval
> Try the new software tools for yourself. Speed compiling, find bugs
> proactively, and fine-tune applications for parallel performance.
> See why Intel Parallel Studio got high marks during beta.
> http://p.sf.net/sfu/intel-sw-dev
> _______________________________________________
> Csound-devel mailing list
> Csound-devel@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/csound-devel
>
>



-- 


Andrés

------------------------------------------------------------------------------
Download Intel® Parallel Studio Eval
Try the new software tools for yourself. Speed compiling, find bugs
proactively, and fine-tune applications for parallel performance.
See why Intel Parallel Studio got high marks during beta.
http://p.sf.net/sfu/intel-sw-dev
_______________________________________________
Csound-devel mailing list
Csound-devel@lists.sourceforge.net
https://lists.

Thanks for the clarification, that's great news, I'll give it a try.

Cheers,
Andrés

On Wed, Mar 3, 2010 at 9:06 AM,   wrote:
> I think you misunderstood me. The situation is this:
>
> 1. The HAL functions in coreaudio are not reentrant; rtcoreaudio.c , which
> is the 'coreaudio' module for Csound uses these functions and is not
> reentrant. The old portaudio library used this as well.
>
> 2. The AudioHAL unit in coreaudio seems to be reentraant; the new portaudio
> library (which I am now bundling with csound) uses this and according to my
> tests is reentrant.
>
> So you can test with the new package (5.12.3) to see if this works for you.
> I am pretty sure it will.
>
> Victor
>
> ----- Original Message -----
> From: Andres Cabrera 
> Date: Wednesday, March 3, 2010 8:55 am
> Subject: Re: [Cs-dev] Found the b****r (was Re: portmidi, still no answers
> in sight)
> To: Developer discussions 
>
>> Hi,
>>
>> I thought you said portaudio would have the same problems as coreaudio
>> if it used the coreaudio interface, so I disabled running multiple
>> instances with portaudio. Would wrapping coreaudio in pulseaudio make
>> coreaudio reentrant?
>>
>> Or should coreaudio work already?
>>
>> Cheers,
>> Andrés
>>
>> On Tue, Mar 2, 2010 at 10:20 PM,
>>  wrote:
>> > What about using portaudio? That should be reentrant too, with
>> the new
>> > version of portaudio I'm distributing.
>> >
>> > Victor
>> >
>> > ----- Original Message -----
>> > From: Andres Cabrera 
>> > Date: Tuesday, March 2, 2010 8:17 pm
>> > Subject: Re: [Cs-dev] Found the b****r (was Re: portmidi,
>> still no answers
>> > in sight)
>> > To: Developer discussions 
>> >
>> >> Hi,
>> >>
>> >> On Tue, Mar 2, 2010 at 6:50 PM,
>> >>  wrote:
>> >> > yes, that seems right to me. The library code is not re-
>> >> entrant, but should work in sequenced calls.
>> >> > I see that Istvan has some code to protect the Pm_Terminate()
>> >> function against being called by an instance before all
>> >> instances are closed, and a similar thing to Pm_Initialize().
>> >> However, the globals in the portmidi library would clash.
>> >> Interesting that they don't on Linux (and Windows?).
>> >> >
>> >>
>> >> I haven't checked whether portmidi is reentrant on linux, since
>> >> I use
>> >> alsa, and that works fine. I'll check soon.
>> >>
>> >> I hope you can find a way to make a reentrant portmidi
>> module, as
>> >> things in QuteCsound are working well now, and you can run several
>> >> instances without a problem on OS X and Linux using jack,
>> which is
>> >> great.
>> >>
>> >> Cheers,
>> >> Andrés
>> >>
>> >> --------------------------------------------------------------
>> ---
>> >> -------------
>> >> Download Intel® Parallel Studio Eval
>> >> Try the new software tools for yourself. Speed compiling,
>> find bugs
>> >> proactively, and fine-tune applications for parallel performance.
>> >> See why Intel Parallel Studio got high marks during beta.
>> >> http://p.sf.net/sfu/intel-sw-dev
>> >> _______________________________________________
>> >> Csound-devel mailing list
>> >> Csound-devel@lists.sourceforge.net
>> >> https://lists.sourceforge.net/lists/listinfo/csound-devel
>> >
>> > Dr Victor Lazzarini, Senior Lecturer, Dept. of Music,
>> > National University of Ireland, Maynooth
>> >
>> > ---------------------------------------------------------------
>> ---------------
>> > Download Intel® Parallel Studio Eval
>> > Try the new software tools for yourself. Speed compiling, find bugs
>> > proactively, and fine-tune applications for parallel performance.
>> > See why Intel Parallel Studio got high marks during beta.
>> > http://p.sf.net/sfu/intel-sw-dev
>> > _______________________________________________
>> > Csound-devel mailing list
>> > Csound-devel@lists.sourceforge.net
>> > https://lists.sourceforge.net/lists/listinfo/csound-devel
>> >
>> >
>>
>>
>>
>> --
>>
>>
>> Andrés
>>
>> -----------------------------------------------------------------
>> -------------
>> Download Intel® Parallel Studio Eval
>> Try the new software tools for yourself. Speed compiling, find bugs
>> proactively, and fine-tune applications for parallel performance.
>> See why Intel Parallel Studio got high marks during beta.
>> http://p.sf.net/sfu/intel-sw-dev
>> _______________________________________________
>> Csound-devel mailing list
>> Csound-devel@lists.sourceforge.net
>> https://lists.sourceforge.net/lists/listinfo/csound-devel
>
> Dr Victor Lazzarini, Senior Lecturer, Dept. of Music,
> National University of Ireland, Maynooth
>
> ------------------------------------------------------------------------------
> Download Intel® Parallel Studio Eval
> Try the new software tools for yourself. Speed compiling, find bugs
> proactively, and fine-tune applications for parallel performance.
> See why Intel Parallel Studio got high marks during beta.
> http://p.sf.net/sfu/intel-sw-dev
> _______________________________________________
> Csound-devel mailing list
> Csound-devel@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/csound-devel
>
>



-- 


Andrés

------------------------------------------------------------------------------
Download Intel® Parallel Studio Eval
Try the new software tools for yourself. Speed compiling, find bugs
proactively, and fine-tune applications for parallel performance.
See why Intel Parallel Studio got high marks during beta.
http://p.sf.net/sfu/intel-sw-dev
_______________________________________________
Csound-devel mailing list
Csound-devel@lists.sourceforge.net
https:

Hi,

Yes, portaudio is now reentrant on OS X.

Thanks very much for this.

Cheers,
Andrés

On Wed, Mar 3, 2010 at 9:14 AM, Andres Cabrera  wrote:
> Thanks for the clarification, that's great news, I'll give it a try.
>
> Cheers,
> Andrés
>
> On Wed, Mar 3, 2010 at 9:06 AM,   wrote:
>> I think you misunderstood me. The situation is this:
>>
>> 1. The HAL functions in coreaudio are not reentrant; rtcoreaudio.c , which
>> is the 'coreaudio' module for Csound uses these functions and is not
>> reentrant. The old portaudio library used this as well.
>>
>> 2. The AudioHAL unit in coreaudio seems to be reentraant; the new portaudio
>> library (which I am now bundling with csound) uses this and according to my
>> tests is reentrant.
>>
>> So you can test with the new package (5.12.3) to see if this works for you.
>> I am pretty sure it will.
>>
>> Victor
>>
>> ----- Original Message -----
>> From: Andres Cabrera 
>> Date: Wednesday, March 3, 2010 8:55 am
>> Subject: Re: [Cs-dev] Found the b****r (was Re: portmidi, still no answers
>> in sight)
>> To: Developer discussions 
>>
>>> Hi,
>>>
>>> I thought you said portaudio would have the same problems as coreaudio
>>> if it used the coreaudio interface, so I disabled running multiple
>>> instances with portaudio. Would wrapping coreaudio in pulseaudio make
>>> coreaudio reentrant?
>>>
>>> Or should coreaudio work already?
>>>
>>> Cheers,
>>> Andrés
>>>
>>> On Tue, Mar 2, 2010 at 10:20 PM,
>>>  wrote:
>>> > What about using portaudio? That should be reentrant too, with
>>> the new
>>> > version of portaudio I'm distributing.
>>> >
>>> > Victor
>>> >
>>> > ----- Original Message -----
>>> > From: Andres Cabrera 
>>> > Date: Tuesday, March 2, 2010 8:17 pm
>>> > Subject: Re: [Cs-dev] Found the b****r (was Re: portmidi,
>>> still no answers
>>> > in sight)
>>> > To: Developer discussions 
>>> >
>>> >> Hi,
>>> >>
>>> >> On Tue, Mar 2, 2010 at 6:50 PM,
>>> >>  wrote:
>>> >> > yes, that seems right to me. The library code is not re-
>>> >> entrant, but should work in sequenced calls.
>>> >> > I see that Istvan has some code to protect the Pm_Terminate()
>>> >> function against being called by an instance before all
>>> >> instances are closed, and a similar thing to Pm_Initialize().
>>> >> However, the globals in the portmidi library would clash.
>>> >> Interesting that they don't on Linux (and Windows?).
>>> >> >
>>> >>
>>> >> I haven't checked whether portmidi is reentrant on linux, since
>>> >> I use
>>> >> alsa, and that works fine. I'll check soon.
>>> >>
>>> >> I hope you can find a way to make a reentrant portmidi
>>> module, as
>>> >> things in QuteCsound are working well now, and you can run several
>>> >> instances without a problem on OS X and Linux using jack,
>>> which is
>>> >> great.
>>> >>
>>> >> Cheers,
>>> >> Andrés
>>> >>
>>> >> --------------------------------------------------------------
>>> ---
>>> >> -------------
>>> >> Download Intel® Parallel Studio Eval
>>> >> Try the new software tools for yourself. Speed compiling,
>>> find bugs
>>> >> proactively, and fine-tune applications for parallel performance.
>>> >> See why Intel Parallel Studio got high marks during beta.
>>> >> http://p.sf.net/sfu/intel-sw-dev
>>> >> _______________________________________________
>>> >> Csound-devel mailing list
>>> >> Csound-devel@lists.sourceforge.net
>>> >> https://lists.sourceforge.net/lists/listinfo/csound-devel
>>> >
>>> > Dr Victor Lazzarini, Senior Lecturer, Dept. of Music,
>>> > National University of Ireland, Maynooth
>>> >
>>> > ---------------------------------------------------------------
>>> ---------------
>>> > Download Intel® Parallel Studio Eval
>>> > Try the new software tools for yourself. Speed compiling, find bugs
>>> > proactively, and fine-tune applications for parallel performance.
>>> > See why Intel Parallel Studio got high marks during beta.
>>> > http://p.sf.net/sfu/intel-sw-dev
>>> > _______________________________________________
>>> > Csound-devel mailing list
>>> > Csound-devel@lists.sourceforge.net
>>> > https://lists.sourceforge.net/lists/listinfo/csound-devel
>>> >
>>> >
>>>
>>>
>>>
>>> --
>>>
>>>
>>> Andrés
>>>
>>> -----------------------------------------------------------------
>>> -------------
>>> Download Intel® Parallel Studio Eval
>>> Try the new software tools for yourself. Speed compiling, find bugs
>>> proactively, and fine-tune applications for parallel performance.
>>> See why Intel Parallel Studio got high marks during beta.
>>> http://p.sf.net/sfu/intel-sw-dev
>>> _______________________________________________
>>> Csound-devel mailing list
>>> Csound-devel@lists.sourceforge.net
>>> https://lists.sourceforge.net/lists/listinfo/csound-devel
>>
>> Dr Victor Lazzarini, Senior Lecturer, Dept. of Music,
>> National University of Ireland, Maynooth
>>
>> ------------------------------------------------------------------------------
>> Download Intel® Parallel Studio Eval
>> Try the new software tools for yourself. Speed compiling, find bugs
>> proactively, and fine-tune applications for parallel performance.
>> See why Intel Parallel Studio got high marks during beta.
>> http://p.sf.net/sfu/intel-sw-dev
>> _______________________________________________
>> Csound-devel mailing list
>> Csound-devel@lists.sourceforge.net
>> https://lists.sourceforge.net/lists/listinfo/csound-devel
>>
>>
>
>
>
> --
>
>
> Andrés
>



-- 


Andrés

------------------------------------------------------------------------------
Download Intel® Parallel Studio Eval
Try the new software tools for yourself. Speed compiling, find bugs
proactively, and fine-tune applications for parallel performance.
See why Intel Parallel Studio got high marks during beta.
http://p.sf.net/sfu/intel-sw-dev
_______________________________________________
Csound-devel mailing list
Csound-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/csound-dev