subject:"Re\: \[Ekiga\-devel\-list\] 1st working SVN build Ekiga on FreeBSD"

Re: [Ekiga-devel-list] 1st working SVN build Ekiga on FreeBSD

2008-04-07 Thread Matthias Apitz

El día Monday, April 07, 2008 a las 01:34:30PM +0200, Damien Sandras escribió:

> > 2008/04/07 13:24:37.125   0:45.284   Media Patch:0x2bc60a00 Debug: 
> > src/opal/patch.cxx: PThread::Sleep(5)
> > 2008/04/07 13:24:37.126   0:45.285   Media Patch:0x2bc60800 Debug: 
> > src/opal/patch.cxx: PThread::Sleep(5)
> > 2008/04/07 13:24:37.135   0:45.294   Media Patch:0x29bd5100 Debug: 
> > src/opal/patch.cxx: PThread::Sleep(5)
> > 2008/04/07 13:24:37.156   0:45.315   Media Patch:0x29bd5100 Debug: 
> > src/opal/patch.cxx: PThread::Sleep(5)
> > 2008/04/07 13:24:37.158   0:45.317   Media Patch:0x2bc60800 Debug: 
> > src/opal/patch.cxx: PThread::Sleep(5)
> 
> I think the question was :
> - does that fix the problem you reported (connection never reaching the
> Established state) ?

sorry, I forgot to say: it does noch change that behaviour of never
reaching the Established state;

Emilio
-- 
Matthias Apitz
Manager Technical Support - OCLC GmbH
Gruenwalder Weg 28g - 82041 Oberhaching - Germany
t +49-89-61308 351 - f +49-89-61308 399 - m +49-170-4527211
e <[EMAIL PROTECTED]> - w http://www.oclc.org/ http://www.UnixArea.de/
b http://gurucubano.blogspot.com/
Don't top-post, read RFC1855 http://www.faqs.org/rfcs/rfc1855.html
A: Because it messes up the order in which people normally read text.
Q: Why is it such a bad thing?
A: Top-posting.
Q: What is the most annoying thing on Usenet and in e-mail?
___
Ekiga-devel-list mailing list
Ekiga-devel-list@gnome.org
http://mail.gnome.org/mailman/listinfo/ekiga-devel-list

Re: [Ekiga-devel-list] 1st working SVN build Ekiga on FreeBSD

2008-04-07 Thread Damien Sandras

Le lundi 07 avril 2008 à 13:31 +0200, Matthias Apitz a écrit :
> El día Monday, April 07, 2008 a las 12:55:14PM +0200, Matthias Schneider 
> escribió:
> 
> > Hi Emilio,
> > could you try to change  in file opal/src/opal/patch.cxx
> > 
> >   452 if (!isSynchronous)
> >   453   PThread::Sleep(5);
> > 
> > to
> > 
> >   452 //if (!isSynchronous)
> >   453   PThread::Sleep(5);
> > 
> > i.e. that sleep is executed everytime?
> 
> I've changed the part to:
> 
> // Matthias Schneider <[EMAIL PROTECTED]> asked me, Emilio, to comment
> // out the if(...)
> PTRACE(1, "Debug: src/opal/patch.cxx: PThread::Sleep(5)");
> // if (!isSynchronous)
>   PThread::Sleep(5);
> 
> and run 'ekiga --sync -d 1'; as you can see it is passing this place with
> high frequency, but never sleeps:
> 
> 2008/04/07 13:24:37.077   0:45.236   Media Patch:0x2bc60a00 Debug: 
> src/opal/patch.cxx: PThread::Sleep(5)
> 2008/04/07 13:24:37.092   0:45.251   Media Patch:0x29bd5100 Debug: 
> src/opal/patch.cxx: PThread::Sleep(5)
> 2008/04/07 13:24:37.094   0:45.253   Media Patch:0x2bc60800 Debug: 
> src/opal/patch.cxx: PThread::Sleep(5)
> 2008/04/07 13:24:37.113   0:45.272   Media Patch:0x29bd5100 Debug: 
> src/opal/patch.cxx: PThread::Sleep(5)
> 2008/04/07 13:24:37.125   0:45.284   Media Patch:0x2bc60a00 Debug: 
> src/opal/patch.cxx: PThread::Sleep(5)
> 2008/04/07 13:24:37.126   0:45.285   Media Patch:0x2bc60800 Debug: 
> src/opal/patch.cxx: PThread::Sleep(5)
> 2008/04/07 13:24:37.135   0:45.294   Media Patch:0x29bd5100 Debug: 
> src/opal/patch.cxx: PThread::Sleep(5)
> 2008/04/07 13:24:37.156   0:45.315   Media Patch:0x29bd5100 Debug: 
> src/opal/patch.cxx: PThread::Sleep(5)
> 2008/04/07 13:24:37.158   0:45.317   Media Patch:0x2bc60800 Debug: 
> src/opal/patch.cxx: PThread::Sleep(5)

I think the question was :
- does that fix the problem you reported (connection never reaching the
Established state) ?
-- 
 _ Damien Sandras
(o-  
//\Ekiga Softphone : http://www.ekiga.org/
v_/_   NOVACOM : http://www.novacom.be/
   FOSDEM  : http://www.fosdem.org/
   SIP Phone   : sip:[EMAIL PROTECTED]
   

___
Ekiga-devel-list mailing list
Ekiga-devel-list@gnome.org
http://mail.gnome.org/mailman/listinfo/ekiga-devel-list

Re: [Ekiga-devel-list] 1st working SVN build Ekiga on FreeBSD

2008-04-07 Thread Matthias Apitz

El día Monday, April 07, 2008 a las 12:55:14PM +0200, Matthias Schneider 
escribió:

> Hi Emilio,
> could you try to change  in file opal/src/opal/patch.cxx
> 
>   452 if (!isSynchronous)
>   453   PThread::Sleep(5);
> 
> to
> 
>   452 //if (!isSynchronous)
>   453   PThread::Sleep(5);
> 
> i.e. that sleep is executed everytime?

I've changed the part to:

// Matthias Schneider <[EMAIL PROTECTED]> asked me, Emilio, to comment
// out the if(...)
PTRACE(1, "Debug: src/opal/patch.cxx: PThread::Sleep(5)");
// if (!isSynchronous)
  PThread::Sleep(5);

and run 'ekiga --sync -d 1'; as you can see it is passing this place with
high frequency, but never sleeps:

2008/04/07 13:24:37.077   0:45.236   Media Patch:0x2bc60a00 Debug: 
src/opal/patch.cxx: PThread::Sleep(5)
2008/04/07 13:24:37.092   0:45.251   Media Patch:0x29bd5100 Debug: 
src/opal/patch.cxx: PThread::Sleep(5)
2008/04/07 13:24:37.094   0:45.253   Media Patch:0x2bc60800 Debug: 
src/opal/patch.cxx: PThread::Sleep(5)
2008/04/07 13:24:37.113   0:45.272   Media Patch:0x29bd5100 Debug: 
src/opal/patch.cxx: PThread::Sleep(5)
2008/04/07 13:24:37.125   0:45.284   Media Patch:0x2bc60a00 Debug: 
src/opal/patch.cxx: PThread::Sleep(5)
2008/04/07 13:24:37.126   0:45.285   Media Patch:0x2bc60800 Debug: 
src/opal/patch.cxx: PThread::Sleep(5)
2008/04/07 13:24:37.135   0:45.294   Media Patch:0x29bd5100 Debug: 
src/opal/patch.cxx: PThread::Sleep(5)
2008/04/07 13:24:37.156   0:45.315   Media Patch:0x29bd5100 Debug: 
src/opal/patch.cxx: PThread::Sleep(5)
2008/04/07 13:24:37.158   0:45.317   Media Patch:0x2bc60800 Debug: 
src/opal/patch.cxx: PThread::Sleep(5)

Emilio
-- 
Matthias Apitz
Manager Technical Support - OCLC GmbH
Gruenwalder Weg 28g - 82041 Oberhaching - Germany
t +49-89-61308 351 - f +49-89-61308 399 - m +49-170-4527211
e <[EMAIL PROTECTED]> - w http://www.oclc.org/ http://www.UnixArea.de/
b http://gurucubano.blogspot.com/
Don't top-post, read RFC1855 http://www.faqs.org/rfcs/rfc1855.html
A: Because it messes up the order in which people normally read text.
Q: Why is it such a bad thing?
A: Top-posting.
Q: What is the most annoying thing on Usenet and in e-mail?
___
Ekiga-devel-list mailing list
Ekiga-devel-list@gnome.org
http://mail.gnome.org/mailman/listinfo/ekiga-devel-list

Re: [Ekiga-devel-list] 1st working SVN build Ekiga on FreeBSD

2008-04-07 Thread Matthias Schneider

Quoting Matthias Schneider <[EMAIL PROTECTED]>:

> Quoting Matthias Apitz <[EMAIL PROTECTED]>:
>
> > El día Friday, April 04, 2008 a las 08:25:31AM +0200, Matthias Apitz
> > escribió:
> >
> > > Hi Matthias,
> > >
> > > This morning I did two tests to [EMAIL PROTECTED], one with V4L and cam,
> > > the other with MovingLogo as picture:
> > >
> > > in both tests also the video was echoed fine from [EMAIL PROTECTED]; I did
> > > realized this for the 1st time because the video echo starts after the
> > > greeting message 'you're about entering the echo test room...' is over;
> > > maybe in the past I never awaited the end of the greeting (or never set
> > > the View to 'picture-in-picture'), anyway: video is fine;
> > >
> > > but in the call with V4L the status message in the main screen stays as
> > > 'Standby', while in the call with MovingLogo it switches to 'Connected
> > > with 500' and the call duration is displayed and updated correctly as
> > > time runs;
> > >
> > > I've inserted some Debug: statements in to the Ekiga code, so they're
> > > not in your code in SVN but the Debug: messages says in which method we
> > > are; here it comes, 1st for the call with V4L and 2nd with MovingLogo;
> > > somehow the call in V4L mode does not reach the connected state and the
> > > HANGUP-button does not work because the X11-call-back function of this
> does
> > > no have the call-pointer in the struct mw->current_call; and this is
> > > perhaps because Opal::Call::OnEstablished() is not passed;
> >
> > I've spent some time to take a closer look at the '-d 3' log files of a
> > call to [EMAIL PROTECTED] using V4L or MovingLogo; it seems that all is fine
> > and the only missing piece is that the call in V4L mode in the Ekiga
> > part gets marked as ESTABLISHED:
> >
> >
> > $ egrep 'PCSS|connection established' d3log-for-MovingLogo
> > 2008/04/04 14:31:45.557   2:41.862  GMURLHandler:0x2be77f00 PCSS
> > Outgoing
> call
> > routed to sip:[EMAIL PROTECTED] for Call[1]-EP[/dev/dsp0]
> > 2008/04/04 14:31:46.346   2:42.650Aggregator:0x29b03900 PCSS
> > Adding
> > filters to patch
> > 2008/04/04 14:31:46.355   2:42.659Aggregator:0x29b03900 PCSS
> > Adding
> > filters to patch
> > 2008/04/04 14:31:46.401   2:42.705
> > Aggregator:0x29b03900   PCSSSetConnected()
> > 2008/04/04 14:31:46.401   2:42.705
> > Aggregator:0x29b03900   PCSSSetConnected() mediaStreams.IsEmpty? -->
> > phase=EstablishedPhase
> > 2008/04/04 14:31:46.401   2:42.705
> > Aggregator:0x29b03900   PCSSSetConnected() calling OnEstablished()
> > 2008/04/04 14:31:46.419   2:42.723Aggregator:0x29b03900 
> > GMPCSSEndpoint
> > PCSS connection established
> > 2008/04/04 14:31:46.420   2:42.724Aggregator:0x29b03900 
> > GMSIPEndpoint
> SIP
> > connection established
> >
> >
> > $ egrep 'PCSS|connection established' d3log-for-V4L
> > 2008/04/04 14:32:27.649   0:21.351  GMURLHandler:0x29bd6300 PCSS
> > Outgoing
> call
> > routed to sip:[EMAIL PROTECTED] for Call[1]-EP[/dev/dsp0]
> > 2008/04/04 14:32:28.162   0:21.865Aggregator:0x2bc67b00 PCSS
> > Adding
> > filters to patch
> > 2008/04/04 14:32:28.170   0:21.872Aggregator:0x2bc67b00 PCSS
> > Adding
> > filters to patch
> > 2008/04/04 14:32:29.503   0:23.205
> > Aggregator:0x2bc67b00   PCSSSetConnected()
> > 2008/04/04 14:32:29.503   0:23.205
> > Aggregator:0x2bc67b00   PCSSSetConnected() mediaStreams.IsEmpty? -->
> > phase=EstablishedPhase
> > 2008/04/04 14:32:29.503   0:23.205
> > Aggregator:0x2bc67b00   PCSSSetConnected() calling OnEstablished()
> >
> > I do not understand why 'PCSSSetConnected()' is reached in both
> > cases, but 'GMPCSSEndpoint   PCSS connection established' only with
> > MovingLogo :-((
> >
> > some ideas?
> >
> > btw: is there some document describing the parts of Ekiga for a new
> > developer who is willing to understand the code, but a bit lost at the
> > beginning in all the classes and pieces?
> >
> > have a nice weekend
> >
> > Emilio
> >
> > --
> > Matthias Apitz
> >
>
>
> Hi Emilio,
> in order to properly track this issue, I have created
>
> http://bugzilla.gnome.org/show_bug.cgi?id=526654
>
> feel free to add additional information.
>
> Matthias
>
> p.s. Sorry no reald dev. documentation exists yet, but will probably be
> created
> during the beta phase. If you are interested in some specific parts of Ekiga,
> maybe we could start with them...
>
Hi Emilio,
could you try to change  in file opal/src/opal/patch.cxx

  452 if (!isSynchronous)
  453   PThread::Sleep(5);

to

  452 //if (!isSynchronous)
  453   PThread::Sleep(5);

i.e. that sleep is executed everytime?

Matthias




This message was sent using IMP, the Internet Messaging Program.
___
Ekiga-devel-list mailing list

Re: [Ekiga-devel-list] 1st working SVN build Ekiga on FreeBSD

2008-04-06 Thread Matthias Schneider

Quoting Matthias Apitz <[EMAIL PROTECTED]>:

> El día Friday, April 04, 2008 a las 08:25:31AM +0200, Matthias Apitz
> escribió:
>
> > Hi Matthias,
> >
> > This morning I did two tests to [EMAIL PROTECTED], one with V4L and cam,
> > the other with MovingLogo as picture:
> >
> > in both tests also the video was echoed fine from [EMAIL PROTECTED]; I did
> > realized this for the 1st time because the video echo starts after the
> > greeting message 'you're about entering the echo test room...' is over;
> > maybe in the past I never awaited the end of the greeting (or never set
> > the View to 'picture-in-picture'), anyway: video is fine;
> >
> > but in the call with V4L the status message in the main screen stays as
> > 'Standby', while in the call with MovingLogo it switches to 'Connected
> > with 500' and the call duration is displayed and updated correctly as
> > time runs;
> >
> > I've inserted some Debug: statements in to the Ekiga code, so they're
> > not in your code in SVN but the Debug: messages says in which method we
> > are; here it comes, 1st for the call with V4L and 2nd with MovingLogo;
> > somehow the call in V4L mode does not reach the connected state and the
> > HANGUP-button does not work because the X11-call-back function of this does
> > no have the call-pointer in the struct mw->current_call; and this is
> > perhaps because Opal::Call::OnEstablished() is not passed;
>
> I've spent some time to take a closer look at the '-d 3' log files of a
> call to [EMAIL PROTECTED] using V4L or MovingLogo; it seems that all is fine
> and the only missing piece is that the call in V4L mode in the Ekiga
> part gets marked as ESTABLISHED:
>
>
> $ egrep 'PCSS|connection established' d3log-for-MovingLogo
> 2008/04/04 14:31:45.557 2:41.862  GMURLHandler:0x2be77f00 PCSS
> Outgoing call
> routed to sip:[EMAIL PROTECTED] for Call[1]-EP[/dev/dsp0]
> 2008/04/04 14:31:46.346 2:42.650Aggregator:0x29b03900 PCSS
> Adding
> filters to patch
> 2008/04/04 14:31:46.355 2:42.659Aggregator:0x29b03900 PCSS
> Adding
> filters to patch
> 2008/04/04 14:31:46.401 2:42.705
> Aggregator:0x29b03900 PCSSSetConnected()
> 2008/04/04 14:31:46.401 2:42.705
> Aggregator:0x29b03900 PCSSSetConnected() mediaStreams.IsEmpty? -->
> phase=EstablishedPhase
> 2008/04/04 14:31:46.401 2:42.705
> Aggregator:0x29b03900 PCSSSetConnected() calling OnEstablished()
> 2008/04/04 14:31:46.419 2:42.723Aggregator:0x29b03900 
> GMPCSSEndpoint
> PCSS connection established
> 2008/04/04 14:31:46.420 2:42.724Aggregator:0x29b03900 
> GMSIPEndpointSIP
> connection established
>
>
> $ egrep 'PCSS|connection established' d3log-for-V4L
> 2008/04/04 14:32:27.649 0:21.351  GMURLHandler:0x29bd6300 PCSS
> Outgoing call
> routed to sip:[EMAIL PROTECTED] for Call[1]-EP[/dev/dsp0]
> 2008/04/04 14:32:28.162 0:21.865Aggregator:0x2bc67b00 PCSS
> Adding
> filters to patch
> 2008/04/04 14:32:28.170 0:21.872Aggregator:0x2bc67b00 PCSS
> Adding
> filters to patch
> 2008/04/04 14:32:29.503 0:23.205
> Aggregator:0x2bc67b00 PCSSSetConnected()
> 2008/04/04 14:32:29.503 0:23.205
> Aggregator:0x2bc67b00 PCSSSetConnected() mediaStreams.IsEmpty? -->
> phase=EstablishedPhase
> 2008/04/04 14:32:29.503 0:23.205
> Aggregator:0x2bc67b00 PCSSSetConnected() calling OnEstablished()
>
> I do not understand why 'PCSSSetConnected()' is reached in both
> cases, but 'GMPCSSEndpoint   PCSS connection established' only with
> MovingLogo :-((
>
> some ideas?
>
> btw: is there some document describing the parts of Ekiga for a new
> developer who is willing to understand the code, but a bit lost at the
> beginning in all the classes and pieces?
>
> have a nice weekend
>
>   Emilio
>
> --
> Matthias Apitz
>


Hi Emilio,
in order to properly track this issue, I have created

http://bugzilla.gnome.org/show_bug.cgi?id=526654

feel free to add additional information.

Matthias

p.s. Sorry no reald dev. documentation exists yet, but will probably be created
during the beta phase. If you are interested in some specific parts of Ekiga,
maybe we could start with them...


This message was sent using IMP, the Internet Messaging Program.
___
Ekiga-devel-list mailing list
Ekiga-devel-list@gnome.org
http://mail.gnome.org/mailman/listinfo/ekiga-devel-list

Re: [Ekiga-devel-list] 1st working SVN build Ekiga on FreeBSD

2008-04-04 Thread Matthias Apitz

El día Friday, April 04, 2008 a las 08:25:31AM +0200, Matthias Apitz escribió:

> Hi Matthias,
> 
> This morning I did two tests to [EMAIL PROTECTED], one with V4L and cam,
> the other with MovingLogo as picture:
> 
> in both tests also the video was echoed fine from [EMAIL PROTECTED]; I did
> realized this for the 1st time because the video echo starts after the
> greeting message 'you're about entering the echo test room...' is over;
> maybe in the past I never awaited the end of the greeting (or never set
> the View to 'picture-in-picture'), anyway: video is fine;
> 
> but in the call with V4L the status message in the main screen stays as
> 'Standby', while in the call with MovingLogo it switches to 'Connected
> with 500' and the call duration is displayed and updated correctly as
> time runs;
> 
> I've inserted some Debug: statements in to the Ekiga code, so they're
> not in your code in SVN but the Debug: messages says in which method we
> are; here it comes, 1st for the call with V4L and 2nd with MovingLogo;
> somehow the call in V4L mode does not reach the connected state and the
> HANGUP-button does not work because the X11-call-back function of this does
> no have the call-pointer in the struct mw->current_call; and this is
> perhaps because Opal::Call::OnEstablished() is not passed;

I've spent some time to take a closer look at the '-d 3' log files of a
call to [EMAIL PROTECTED] using V4L or MovingLogo; it seems that all is fine
and the only missing piece is that the call in V4L mode in the Ekiga
part gets marked as ESTABLISHED:


$ egrep 'PCSS|connection established' d3log-for-MovingLogo
2008/04/04 14:31:45.557   2:41.862  GMURLHandler:0x2be77f00 PCSS
Outgoing call routed to sip:[EMAIL PROTECTED] for Call[1]-EP[/dev/dsp0]
2008/04/04 14:31:46.346   2:42.650Aggregator:0x29b03900 PCSSAdding 
filters to patch
2008/04/04 14:31:46.355   2:42.659Aggregator:0x29b03900 PCSSAdding 
filters to patch
2008/04/04 14:31:46.401   2:42.705Aggregator:0x29b03900 PCSS
SetConnected()
2008/04/04 14:31:46.401   2:42.705Aggregator:0x29b03900 PCSS
SetConnected() mediaStreams.IsEmpty? --> phase=EstablishedPhase
2008/04/04 14:31:46.401   2:42.705Aggregator:0x29b03900 PCSS
SetConnected() calling OnEstablished()
2008/04/04 14:31:46.419   2:42.723Aggregator:0x29b03900 GMPCSSEndpoint  
 PCSS connection established
2008/04/04 14:31:46.420   2:42.724Aggregator:0x29b03900 GMSIPEndpoint   
 SIP connection established


$ egrep 'PCSS|connection established' d3log-for-V4L
2008/04/04 14:32:27.649   0:21.351  GMURLHandler:0x29bd6300 PCSS
Outgoing call routed to sip:[EMAIL PROTECTED] for Call[1]-EP[/dev/dsp0]
2008/04/04 14:32:28.162   0:21.865Aggregator:0x2bc67b00 PCSSAdding 
filters to patch
2008/04/04 14:32:28.170   0:21.872Aggregator:0x2bc67b00 PCSSAdding 
filters to patch
2008/04/04 14:32:29.503   0:23.205Aggregator:0x2bc67b00 PCSS
SetConnected()
2008/04/04 14:32:29.503   0:23.205Aggregator:0x2bc67b00 PCSS
SetConnected() mediaStreams.IsEmpty? --> phase=EstablishedPhase
2008/04/04 14:32:29.503   0:23.205Aggregator:0x2bc67b00 PCSS
SetConnected() calling OnEstablished()

I do not understand why 'PCSSSetConnected()' is reached in both
cases, but 'GMPCSSEndpoint   PCSS connection established' only with
MovingLogo :-((

some ideas?

btw: is there some document describing the parts of Ekiga for a new
developer who is willing to understand the code, but a bit lost at the
beginning in all the classes and pieces?

have a nice weekend

Emilio

-- 
Matthias Apitz
___
Ekiga-devel-list mailing list
Ekiga-devel-list@gnome.org
http://mail.gnome.org/mailman/listinfo/ekiga-devel-list

Re: [Ekiga-devel-list] 1st working SVN build Ekiga on FreeBSD

2008-04-04 Thread Matthias Schneider

Quoting Matthias Apitz <[EMAIL PROTECTED]>:

> El día Wednesday, April 02, 2008 a las 06:09:26PM +0200, Matthias Schneider
> escribió:
>
> > Emilio,
> >
> > sorry, thats all I know about H.264 and H.323 and I do not have any
> resources to
> > deal with it any further. If you are really interested you can dig into
> this for
> > yourself and contact Simone Horne (howerver only for questions about OPAL
> and
> > not about Ekiga), there should be posts by him on the OPAL mailing list.
>
> (btw: it is Simon Horne <[EMAIL PROTECTED]> and I contacted him, no
> reply yet);
>
> >
> > However I would like to help you on the video issue, so I would welcome
> > additional info there (preview working?, logo working with 500, etc.)
> >
> > Matthias
>
> Hi Matthias,
>
> This morning I did two tests to [EMAIL PROTECTED], one with V4L and cam,
> the other with MovingLogo as picture:
>
> in both tests also the video was echoed fine from [EMAIL PROTECTED]; I did
> realized this for the 1st time because the video echo starts after the
> greeting message 'you're about entering the echo test room...' is over;
> maybe in the past I never awaited the end of the greeting (or never set
> the View to 'picture-in-picture'), anyway: video is fine;
>
> but in the call with V4L the status message in the main screen stays as
> 'Standby', while in the call with MovingLogo it switches to 'Connected
> with 500' and the call duration is displayed and updated correctly as
> time runs;
>
> I've inserted some Debug: statements in to the Ekiga code, so they're
> not in your code in SVN but the Debug: messages says in which method we
> are; here it comes, 1st for the call with V4L and 2nd with MovingLogo;
> somehow the call in V4L mode does not reach the connected state and the
> HANGUP-button does not work because the X11-call-back function of this does
> no have the call-pointer in the struct mw->current_call; and this is
> perhaps because Opal::Call::OnEstablished() is not passed;
>
>
> call to [EMAIL PROTECTED] with V4L:
>
> Debug:lib/engine/protocol/skel/call-core.cpp CallCore::dial()
> Debug:src/endpoints/opal-call.cpp Opal::Call::Call()
> Debug:src/endpoints/opal-call.cpp Opal::Call::OnOpenMediaStream()
> Debug:StreamType Audio
> Debug:stream_name: PCMU
> Debug:src/endpoints/opal-call.cpp Opal::Call::OnOpenMediaStream()
> Debug:StreamType Audio
> Debug:stream_name: PCMU
> Debug:src/endpoints/opal-call.cpp Opal::Call::OnOpenMediaStream()
> Debug:StreamType Video
> Debug:stream_name: h261
> Debug:src/endpoints/opal-call.cpp Opal::Call::OnOpenMediaStream()
> Debug:StreamType Video
> Debug:stream_name: h261
> CIF MPI = 33
> Frame Time = 3003
> Frame Height = 144
> Temporal Spatial Trade Off = 0
> Max Bit Rate = 4096000
> Clock Rate = 9
> Max Rx Frame Height = 144
> Max Rx Frame Width = 176
> QCIF MPI = 33
> Min Rx Frame Height = 144
> Min Rx Frame Width = 176
> Annex D = 0
> Max Frame Size = 1400
> Target Bit Rate = 80
> Frame Width = 176
>
> [click on HANGUP button]
>
> Debug:toggle_call_cb()
> Debug:mw->current_call is NULL
> gm_connect_button_get_connected... is true
> Debug:now leaving toggle_call_cb
>
> CTRL-C to end ekiga;
>
>
> call to [EMAIL PROTECTED] with MovingLogo:
>
> Debug:lib/engine/protocol/skel/call-core.cpp CallCore::dial()
> Debug:src/endpoints/opal-call.cpp Opal::Call::Call()
> Debug:src/endpoints/opal-call.cpp Opal::Call::OnOpenMediaStream()
> Debug:StreamType Audio
> Debug:stream_name: PCMU
> Debug:src/endpoints/opal-call.cpp Opal::Call::OnOpenMediaStream()
> Debug:StreamType Audio
> Debug:stream_name: PCMU
> Debug:src/endpoints/opal-call.cpp Opal::Call::OnOpenMediaStream()
> Debug:StreamType Video
> Debug:stream_name: h261
> Debug:src/endpoints/opal-call.cpp Opal::Call::OnOpenMediaStream()
> Debug:StreamType Video
> Debug:stream_name: h261
> CIF MPI = 33
> Frame Time = 3003
> Frame Height = 144
> Temporal Spatial Trade Off = 0
> Max Bit Rate = 4096000
> Clock Rate = 9
> Max Rx Frame Height = 144
> Max Rx Frame Width = 176
> QCIF MPI = 33
> Min Rx Frame Height = 144
> Min Rx Frame Width = 176
> Annex D = 0
> Max Frame Size = 1400
> Target Bit Rate = 80
> Frame Width = 176
> Debug:src/endpoints/opal-call.cpp Opal::Call::OnEstablished()
> Debug:src/endpoints/opal-call.cpp Opal::Call::OnEstablished()
> Debug:on_established_call_cb
> Debug:setting mw->current_call = &call;
>
> [click on HANGUP button]
>
> Debug:toggle_call_cb()
> gm_connect_button_get_connected... is true
> Debug:calling hangup_call_cb now
> Debug:hangup_call_cb()
> Debug:src/endpoints/opal-call.cpp Opal::Call::hangup()
> Debug:now leaving toggle_call_cb
>
> The program 'ekiga' received an X Window System error.
> This probably reflects a bug in the program.
> The error was 'BadWindow (invalid Window parameter)'.
>   (Details: serial 127476 error_code 3 request_code 10 minor_code 0)
>   (Note to programmers: normally, X errors are reported asynchronously;
>that is, you will receive the error a while after causing it.
>To debug your p

Re: [Ekiga-devel-list] 1st working SVN build Ekiga on FreeBSD

2008-04-03 Thread Matthias Apitz

El día Wednesday, April 02, 2008 a las 06:09:26PM +0200, Matthias Schneider 
escribió:

> Emilio,
> 
> sorry, thats all I know about H.264 and H.323 and I do not have any resources 
> to
> deal with it any further. If you are really interested you can dig into this 
> for
> yourself and contact Simone Horne (howerver only for questions about OPAL and
> not about Ekiga), there should be posts by him on the OPAL mailing list.

(btw: it is Simon Horne <[EMAIL PROTECTED]> and I contacted him, no
reply yet);

> 
> However I would like to help you on the video issue, so I would welcome
> additional info there (preview working?, logo working with 500, etc.)
> 
> Matthias

Hi Matthias,

This morning I did two tests to [EMAIL PROTECTED], one with V4L and cam,
the other with MovingLogo as picture:

in both tests also the video was echoed fine from [EMAIL PROTECTED]; I did
realized this for the 1st time because the video echo starts after the
greeting message 'you're about entering the echo test room...' is over;
maybe in the past I never awaited the end of the greeting (or never set
the View to 'picture-in-picture'), anyway: video is fine;

but in the call with V4L the status message in the main screen stays as
'Standby', while in the call with MovingLogo it switches to 'Connected
with 500' and the call duration is displayed and updated correctly as
time runs;

I've inserted some Debug: statements in to the Ekiga code, so they're
not in your code in SVN but the Debug: messages says in which method we
are; here it comes, 1st for the call with V4L and 2nd with MovingLogo;
somehow the call in V4L mode does not reach the connected state and the
HANGUP-button does not work because the X11-call-back function of this does
no have the call-pointer in the struct mw->current_call; and this is
perhaps because Opal::Call::OnEstablished() is not passed;


call to [EMAIL PROTECTED] with V4L:

Debug:lib/engine/protocol/skel/call-core.cpp CallCore::dial()
Debug:src/endpoints/opal-call.cpp Opal::Call::Call()
Debug:src/endpoints/opal-call.cpp Opal::Call::OnOpenMediaStream()
Debug:StreamType Audio
Debug:stream_name: PCMU
Debug:src/endpoints/opal-call.cpp Opal::Call::OnOpenMediaStream()
Debug:StreamType Audio
Debug:stream_name: PCMU
Debug:src/endpoints/opal-call.cpp Opal::Call::OnOpenMediaStream()
Debug:StreamType Video
Debug:stream_name: h261
Debug:src/endpoints/opal-call.cpp Opal::Call::OnOpenMediaStream()
Debug:StreamType Video
Debug:stream_name: h261
CIF MPI = 33 
Frame Time = 3003 
Frame Height = 144 
Temporal Spatial Trade Off = 0 
Max Bit Rate = 4096000 
Clock Rate = 9 
Max Rx Frame Height = 144 
Max Rx Frame Width = 176 
QCIF MPI = 33 
Min Rx Frame Height = 144 
Min Rx Frame Width = 176 
Annex D = 0 
Max Frame Size = 1400 
Target Bit Rate = 80 
Frame Width = 176 

[click on HANGUP button]

Debug:toggle_call_cb()
Debug:mw->current_call is NULL
gm_connect_button_get_connected... is true
Debug:now leaving toggle_call_cb

CTRL-C to end ekiga;


call to [EMAIL PROTECTED] with MovingLogo:

Debug:lib/engine/protocol/skel/call-core.cpp CallCore::dial()
Debug:src/endpoints/opal-call.cpp Opal::Call::Call()
Debug:src/endpoints/opal-call.cpp Opal::Call::OnOpenMediaStream()
Debug:StreamType Audio
Debug:stream_name: PCMU
Debug:src/endpoints/opal-call.cpp Opal::Call::OnOpenMediaStream()
Debug:StreamType Audio
Debug:stream_name: PCMU
Debug:src/endpoints/opal-call.cpp Opal::Call::OnOpenMediaStream()
Debug:StreamType Video
Debug:stream_name: h261
Debug:src/endpoints/opal-call.cpp Opal::Call::OnOpenMediaStream()
Debug:StreamType Video
Debug:stream_name: h261
CIF MPI = 33 
Frame Time = 3003 
Frame Height = 144 
Temporal Spatial Trade Off = 0 
Max Bit Rate = 4096000 
Clock Rate = 9 
Max Rx Frame Height = 144 
Max Rx Frame Width = 176 
QCIF MPI = 33 
Min Rx Frame Height = 144 
Min Rx Frame Width = 176 
Annex D = 0 
Max Frame Size = 1400 
Target Bit Rate = 80 
Frame Width = 176 
Debug:src/endpoints/opal-call.cpp Opal::Call::OnEstablished()
Debug:src/endpoints/opal-call.cpp Opal::Call::OnEstablished()
Debug:on_established_call_cb
Debug:setting mw->current_call = &call;

[click on HANGUP button]

Debug:toggle_call_cb()
gm_connect_button_get_connected... is true
Debug:calling hangup_call_cb now
Debug:hangup_call_cb()
Debug:src/endpoints/opal-call.cpp Opal::Call::hangup()
Debug:now leaving toggle_call_cb

The program 'ekiga' received an X Window System error.
This probably reflects a bug in the program.
The error was 'BadWindow (invalid Window parameter)'.
  (Details: serial 127476 error_code 3 request_code 10 minor_code 0)
  (Note to programmers: normally, X errors are reported asynchronously;
   that is, you will receive the error a while after causing it.
   To debug your program, run it with the --sync command line
   option to change this behavior. You can then get a meaningful
   backtrace from your debugger if you break on the gdk_x_error() function.)
Segmentation fault (core dumped)
$

that's all for the moment; comments?

Re: [Ekiga-devel-list] 1st working SVN build Ekiga on FreeBSD

2008-04-03 Thread Matthias Apitz

El día Thursday, April 03, 2008 a las 12:42:45PM +0200, Matthias Schneider 
escribió:

> > no add-onn software (like OPAL, Ekiga) install in FreeBSD stuff below
> > /usr/lib, only below /usr/local; this is the common rule in FreeBSD;
> 
> Hmm, strange, on Linux (as far as I know) stuff that comes with the 
> distribution
> goes to /usr/lib, i.e. opal and ptlib if they come with the distro. Isnt it
> similar on freeBSD?

No, only the core system installs libs into /usr/lib; what you add from
the ports collection install all in /usr/local/... 
things are more clean on FreeBSD :-)

> Actually no harm is done checking in /usr/lib, so I would prefer to keep this
> check without a define...

it's up to the maintainer of the source, i.e. not me, but in FreeBSD it's
definitely the wrong place to look for shared objects of PTLib or OPAL;

Emilio
-- 
Matthias Apitz
___
Ekiga-devel-list mailing list
Ekiga-devel-list@gnome.org
http://mail.gnome.org/mailman/listinfo/ekiga-devel-list

Re: [Ekiga-devel-list] 1st working SVN build Ekiga on FreeBSD

2008-04-03 Thread Matthias Schneider

Quoting Matthias Apitz <[EMAIL PROTECTED]>:

> El día Thursday, April 03, 2008 a las 11:45:08AM +0200, Matthias Schneider
> escribió:
>
> > Hello Emilio,
> > I do not concur with all parts of your patch:
> > - why do you change P_DEFAULT_PLUGIN_DIR? Its not used anyways and should
> be
> > removed -> I will do so
>
> fine with me too;
>
> > - why do you not want to do the check for /usr/lib on freeBSD? This is
> usual
> > procedure for libraries as well. In case it is not found it will be looked
> for
> > in /usr/local/lib. What I can propose is to set the trace level to DEBUG,
> not
> > ERROR for the dir names that are being searched.
>
> no add-onn software (like OPAL, Ekiga) install in FreeBSD stuff below
> /usr/lib, only below /usr/local; this is the common rule in FreeBSD;

Hmm, strange, on Linux (as far as I know) stuff that comes with the distribution
goes to /usr/lib, i.e. opal and ptlib if they come with the distro. Isnt it
similar on freeBSD?

Actually no harm is done checking in /usr/lib, so I would prefer to keep this
check without a define...

> > - You added an if in case a filename with length = 0 should be openend.
> This
> > makes sense if x264 was not detected at compile time. I will add this.
>
> ok;
>
>   Emilio
> --
> Matthias Apitz
>





This message was sent using IMP, the Internet Messaging Program.
___
Ekiga-devel-list mailing list
Ekiga-devel-list@gnome.org
http://mail.gnome.org/mailman/listinfo/ekiga-devel-list

Re: [Ekiga-devel-list] 1st working SVN build Ekiga on FreeBSD

2008-04-03 Thread Matthias Apitz

El día Thursday, April 03, 2008 a las 11:45:08AM +0200, Matthias Schneider 
escribió:

> Hello Emilio,
> I do not concur with all parts of your patch:
> - why do you change P_DEFAULT_PLUGIN_DIR? Its not used anyways and should be
> removed -> I will do so

fine with me too;

> - why do you not want to do the check for /usr/lib on freeBSD? This is usual
> procedure for libraries as well. In case it is not found it will be looked for
> in /usr/local/lib. What I can propose is to set the trace level to DEBUG, not
> ERROR for the dir names that are being searched.

no add-onn software (like OPAL, Ekiga) install in FreeBSD stuff below
/usr/lib, only below /usr/local; this is the common rule in FreeBSD;

> - You added an if in case a filename with length = 0 should be openend. This
> makes sense if x264 was not detected at compile time. I will add this.

ok;

Emilio
-- 
Matthias Apitz
___
Ekiga-devel-list mailing list
Ekiga-devel-list@gnome.org
http://mail.gnome.org/mailman/listinfo/ekiga-devel-list

Re: [Ekiga-devel-list] 1st working SVN build Ekiga on FreeBSD

2008-04-03 Thread Matthias Schneider

Quoting Matthias Apitz <[EMAIL PROTECTED]>:

> El día Thursday, April 03, 2008 a las 10:52:22AM +1100, Craig Southeren
> escribió:
>
> > Matthias Schneider wrote:
> >
> > ..deleted
> >
> > > Emilio,
> > >
> > > sorry, thats all I know about H.264 and H.323 and I do not have any
> resources to
> > > deal with it any further. If you are really interested you can dig into
> this for
> > > yourself and contact Simone Horne (howerver only for questions about OPAL
> and
> > > not about Ekiga), there should be posts by him on the OPAL mailing list.
> > >
> > > However I would like to help you on the video issue, so I would welcome
> > > additional info there (preview working?, logo working with 500, etc.)
>
> Matthias,
> I will continue tomorrow with the tests about that video issue in calls
> to <[EMAIL PROTECTED]> and let you know;
>
> >
> > Not sure Simon would be interested in OPAL, as he is the creator of the
> > h323plus, which is based on OpenH323.
> >
> > But I would be happy to help you with any questions :)
>
> Craig,
> For me (in my work) it would be nice to get this H.264 working in H.323
> because our central VideoConf system does not understand H.261 (bug in
> firmware); that's why I have enabled that define H323_H264_TEST in
> opal/plugins/video/H.264/h264-x264.h, recompiled all stuff including
> ekiga itself, but now it complains about the child proc terminating for
> h264-x264 and dumps core:
>
> $ ekiga -d 3 --sync
> dyna.cxx(98)  H263+   DYNASuccessfully loaded libavcodec.so
> dyna.cxx(326) H263+   DYNASuccessfully loaded libavcodec library 
> and
> verified functions
>h263-1998.cxx(907) H263+   Codec   Enabled
> dyna.cxx(98)  MPEG4   DYNASuccessfully loaded libavcodec.so
> dyna.cxx(326) MPEG4   DYNASuccessfully loaded libavcodec library 
> and
> verified functions
>mpeg4.cxx(2028)MPEG4   Codec   Enabled
> dyna.cxx(98)  H264DYNASuccessfully loaded libavcodec.so
> dyna.cxx(326) H264DYNASuccessfully loaded libavcodec library 
> and
> verified functions
> h264pipe_unix.cxx(295)H264IPC PP: Found GPL process 
> executable in
> /usr/local/lib/ptlib/codecs/video/h264_video_pwplugin_helper
> x264loader_unix.cxx(125)  H264DYNASuccessfully loaded libx264.so
> x264loader_unix.cxx(101)  H264DYNALoader was compiled with x264 
> build 56
> present
> x264loader_unix.cxx(104)  H264DYNASuccessfully loaded libx264 
> library and
> verified functions
> h264pipe_unix.cxx(122)H264IPC PP: Successfully forked child 
> process 19248
> and established communication
>h264-x264.cxx(775) H264Codec   Enabled
> h264helper_unix.cxx(68)   H264IPC CP: Terminating
> Segmentation fault (core dumped)
>
> if you have an idea where to look, your helping hand is welcome;
>
> Damien,
> I'm attaching some smaller changes for OPAL dealing with places where to
> pick up objects in FreeBSD;
>
>   Emilio
> --
> Matthias Apitz
>
Hello Emilio,
I do not concur with all parts of your patch:
- why do you change P_DEFAULT_PLUGIN_DIR? Its not used anyways and should be
removed -> I will do so
- why do you not want to do the check for /usr/lib on freeBSD? This is usual
procedure for libraries as well. In case it is not found it will be looked for
in /usr/local/lib. What I can propose is to set the trace level to DEBUG, not
ERROR for the dir names that are being searched.
- You added an if in case a filename with length = 0 should be openend. This
makes sense if x264 was not detected at compile time. I will add this.

Matthias




This message was sent using IMP, the Internet Messaging Program.
___
Ekiga-devel-list mailing list
Ekiga-devel-list@gnome.org
http://mail.gnome.org/mailman/listinfo/ekiga-devel-list

Re: [Ekiga-devel-list] 1st working SVN build Ekiga on FreeBSD

2008-04-03 Thread Matthias Apitz

El día Thursday, April 03, 2008 a las 10:52:22AM +1100, Craig Southeren 
escribió:

> Matthias Schneider wrote:
> 
> ..deleted
> 
> > Emilio,
> > 
> > sorry, thats all I know about H.264 and H.323 and I do not have any 
> > resources to
> > deal with it any further. If you are really interested you can dig into 
> > this for
> > yourself and contact Simone Horne (howerver only for questions about OPAL 
> > and
> > not about Ekiga), there should be posts by him on the OPAL mailing list.
> > 
> > However I would like to help you on the video issue, so I would welcome
> > additional info there (preview working?, logo working with 500, etc.)

Matthias,
I will continue tomorrow with the tests about that video issue in calls
to <[EMAIL PROTECTED]> and let you know;

> 
> Not sure Simon would be interested in OPAL, as he is the creator of the 
> h323plus, which is based on OpenH323.
> 
> But I would be happy to help you with any questions :)

Craig,
For me (in my work) it would be nice to get this H.264 working in H.323
because our central VideoConf system does not understand H.261 (bug in
firmware); that's why I have enabled that define H323_H264_TEST in
opal/plugins/video/H.264/h264-x264.h, recompiled all stuff including
ekiga itself, but now it complains about the child proc terminating for
h264-x264 and dumps core:

$ ekiga -d 3 --sync
dyna.cxx(98)H263+   DYNASuccessfully loaded libavcodec.so
dyna.cxx(326)   H263+   DYNASuccessfully loaded libavcodec library 
and verified functions
   h263-1998.cxx(907)   H263+   Codec   Enabled
dyna.cxx(98)MPEG4   DYNASuccessfully loaded libavcodec.so
dyna.cxx(326)   MPEG4   DYNASuccessfully loaded libavcodec library 
and verified functions
   mpeg4.cxx(2028)  MPEG4   Codec   Enabled
dyna.cxx(98)H264DYNASuccessfully loaded libavcodec.so
dyna.cxx(326)   H264DYNASuccessfully loaded libavcodec library 
and verified functions
h264pipe_unix.cxx(295)  H264IPC PP: Found GPL process executable in  
/usr/local/lib/ptlib/codecs/video/h264_video_pwplugin_helper
x264loader_unix.cxx(125)H264DYNASuccessfully loaded libx264.so
x264loader_unix.cxx(101)H264DYNALoader was compiled with x264 
build 56 present
x264loader_unix.cxx(104)H264DYNASuccessfully loaded libx264 
library and verified functions
h264pipe_unix.cxx(122)  H264IPC PP: Successfully forked child process 
19248 and established communication
   h264-x264.cxx(775)   H264Codec   Enabled
h264helper_unix.cxx(68) H264IPC CP: Terminating
Segmentation fault (core dumped)

if you have an idea where to look, your helping hand is welcome;

Damien,
I'm attaching some smaller changes for OPAL dealing with places where to
pick up objects in FreeBSD;

Emilio
-- 
Matthias Apitz
*** plugins/video/H.264/gpl/x264loader_unix.cxx	2008-04-03 09:36:12.0 +0200
--- plugins/video/H.264/gpl/x264loader_unix.cxx.orig	2008-03-27 14:34:39.0 +0100
***
*** 109,120 
  
  bool X264Library::Open(const char *name)
  {
-   if ( strlen(name) == 0 )
-   return false;
_dynamicLibrary = dlopen(name, RTLD_NOW);
  
if (_dynamicLibrary == NULL) {
! char * error = (char *) dlerror();
  if (error != NULL)
  TRACE(1, "H264\tDYNA\tError loading " << name << " - " << error)
else
--- 109,118 
  
  bool X264Library::Open(const char *name)
  {
_dynamicLibrary = dlopen(name, RTLD_NOW);
  
if (_dynamicLibrary == NULL) {
! char * error = dlerror();
  if (error != NULL)
  TRACE(1, "H264\tDYNA\tError loading " << name << " - " << error)
else
*** plugins/video/H.264/h264pipe_unix.cxx	2008-04-03 09:55:58.0 +0200
--- plugins/video/H.264/h264pipe_unix.cxx.orig	2008-04-03 09:45:34.0 +0200
***
*** 34,44 
  
  #define HAS_MKFIFO 1
  #define GPL_PROCESS_FILENAME "ptlib/codecs/video/h264_video_pwplugin_helper"
! #ifdef P_FREEBSD
!   #define P_DEFAULT_PLUGIN_DIR "/usr/local"
! #else
!   #define P_DEFAULT_PLUGIN_DIR "/usr/lib/ptlib"
! #endif
  #define DIR_SEPERATOR "/"
  #define DIR_TOKENISER ":"
  
--- 34,40 
  
  #define HAS_MKFIFO 1
  #define GPL_PROCESS_FILENAME "ptlib/codecs/video/h264_video_pwplugin_helper"
! #define P_DEFAULT_PLUGIN_DIR "/usr/lib/ptlib"
  #define DIR_SEPERATOR "/"
  #define DIR_TOKENISER ":"
  
***
*** 266,275 
  }
}
  
- #ifndef P_FREEBSD
if (checkGplProcessExists("/usr/lib")) 
  return true;
- #endif
  
if (checkGplProcessExists("/usr/local/lib")) 
  return true;
--- 262,269 
___
Ekiga-devel-list mailing list
Ekiga-devel-list@gnome.org
http://mail.gnome.org/mailman/listinfo/ekiga-devel-list

Re: [Ekiga-devel-list] 1st working SVN build Ekiga on FreeBSD

2008-04-02 Thread Craig Southeren

Matthias Schneider wrote:

..deleted

> Emilio,
> 
> sorry, thats all I know about H.264 and H.323 and I do not have any resources 
> to
> deal with it any further. If you are really interested you can dig into this 
> for
> yourself and contact Simone Horne (howerver only for questions about OPAL and
> not about Ekiga), there should be posts by him on the OPAL mailing list.
> 
> However I would like to help you on the video issue, so I would welcome
> additional info there (preview working?, logo working with 500, etc.)

Not sure Simon would be interested in OPAL, as he is the creator of the 
h323plus, which is based on OpenH323.

But I would be happy to help you with any questions :)

Craig

---
  Craig Southeren  Post Increment – VoIP Consulting and Software
  [EMAIL PROTECTED]   www.postincrement.com.au

  Phone:  +61 243654666  ICQ: #86852844
  Fax:+61 243656905  MSN: [EMAIL PROTECTED]
  Mobile: +61 417231046  Jabber: [EMAIL PROTECTED]

  "Science is the poetry of reality." Richard Dawkins

___
Ekiga-devel-list mailing list
Ekiga-devel-list@gnome.org
http://mail.gnome.org/mailman/listinfo/ekiga-devel-list

Re: [Ekiga-devel-list] 1st working SVN build Ekiga on FreeBSD

2008-04-02 Thread Matthias Schneider

Quoting Matthias Apitz <[EMAIL PROTECTED]>:

> El día Wednesday, April 02, 2008 a las 03:43:55PM +0200, Matthias Apitz
> escribió:
>
> > I did H.323 tests against someone using Netmeeting (which worked fine)
> > and against our VSX7000 VC system (in which at least I can see what the
> > VSX7000 is sending as video); both tests have be done inside our VPN;
> > I'll check with tcpdump if in connection to [EMAIL PROTECTED] via Internet
> > it is not our firewall that disables the receiving...
> >
> > > > do you have any plans about H.264 in H.323?
> > > None from my side but there is an experimental implmenetation by Simone
> Horne
> > > you can activate via a simple #define.
> > > Matthias
>
> Matthias, I have found the place and activated the #define H323_H264_TEST;
>
> now the compilation stops with:
> ...
> gmake[3]: Entering directory `/usr/home/guru/Ekiga/opal/plugins/video/H.264'
> g++ -I../../../include -DLICENCE_MPL  -DWITH_RFC_COMPLIANT_DEFAULTS
> -I../common -fPIC  -DH323_H264_TEST -DP_FREEBSD -I/usr/local/include -c
> h264-x264.cxx -o obj/h264-x264.o
> In file included from h264-x264.h:61,
>  from h264-x264.cxx:38:
> /usr/local/include/ffmpeg/avcodec.h:2248: warning: 'ImgReSampleContext' is
> deprecated (declared at /usr/local/include/ffmpeg/avcodec.h:2242)
> /usr/local/include/ffmpeg/avcodec.h:2258: warning: 'ImgReSampleContext' is
> deprecated (declared at /usr/local/include/ffmpeg/avcodec.h:2242)
> In file included from h264-x264.cxx:38:
> h264-x264.h:379: error: braces around scalar initializer for type 'signed
> char:1'
> h264-x264.h:379: error: braces around scalar initializer for type 'signed
> char:1'
> h264-x264.h:390: error: braces around scalar initializer for type 'signed
> char:1'
> h264-x264.h:390: error: braces around scalar initializer for type 'signed
> char:1'
> h264-x264.h:402: error: braces around scalar initializer for type 'signed
> char:1'
> h264-x264.h:402: error: braces around scalar initializer for type 'signed
> char:1'
> h264-x264.h:414: error: braces around scalar initializer for type 'signed
> char:1'
> h264-x264.h:414: error: braces around scalar initializer for type 'signed
> char:1'
> h264-x264.h:426: error: braces around scalar initializer for type 'signed
> char:1'
> h264-x264.h:426: error: braces around scalar initializer for type 'signed
> char:1'
>
> the problem is the macro DECLARE_GENERIC_OPTIONS(...) in h264-x264.h
> and a patch is attached, which let it compile;
>
> but: in Ekiga it shows H.264 again only for SIP and not for H.323; do I
> have to activate something else?
>
>   Emilio
> --
> Matthias Apitz
>
Emilio,

sorry, thats all I know about H.264 and H.323 and I do not have any resources to
deal with it any further. If you are really interested you can dig into this for
yourself and contact Simone Horne (howerver only for questions about OPAL and
not about Ekiga), there should be posts by him on the OPAL mailing list.

However I would like to help you on the video issue, so I would welcome
additional info there (preview working?, logo working with 500, etc.)

Matthias




This message was sent using IMP, the Internet Messaging Program.
___
Ekiga-devel-list mailing list
Ekiga-devel-list@gnome.org
http://mail.gnome.org/mailman/listinfo/ekiga-devel-list

Re: [Ekiga-devel-list] 1st working SVN build Ekiga on FreeBSD

2008-04-02 Thread Matthias Apitz

El día Wednesday, April 02, 2008 a las 03:43:55PM +0200, Matthias Apitz 
escribió:

> I did H.323 tests against someone using Netmeeting (which worked fine)
> and against our VSX7000 VC system (in which at least I can see what the
> VSX7000 is sending as video); both tests have be done inside our VPN;
> I'll check with tcpdump if in connection to [EMAIL PROTECTED] via Internet
> it is not our firewall that disables the receiving...
> 
> > > do you have any plans about H.264 in H.323?
> > None from my side but there is an experimental implmenetation by Simone 
> > Horne
> > you can activate via a simple #define.
> > Matthias

Matthias, I have found the place and activated the #define H323_H264_TEST;

now the compilation stops with:
...
gmake[3]: Entering directory `/usr/home/guru/Ekiga/opal/plugins/video/H.264'
g++ -I../../../include -DLICENCE_MPL  -DWITH_RFC_COMPLIANT_DEFAULTS -I../common 
-fPIC  -DH323_H264_TEST -DP_FREEBSD -I/usr/local/include -c h264-x264.cxx -o 
obj/h264-x264.o
In file included from h264-x264.h:61,
 from h264-x264.cxx:38:
/usr/local/include/ffmpeg/avcodec.h:2248: warning: 'ImgReSampleContext' is 
deprecated (declared at /usr/local/include/ffmpeg/avcodec.h:2242)
/usr/local/include/ffmpeg/avcodec.h:2258: warning: 'ImgReSampleContext' is 
deprecated (declared at /usr/local/include/ffmpeg/avcodec.h:2242)
In file included from h264-x264.cxx:38:
h264-x264.h:379: error: braces around scalar initializer for type 'signed 
char:1'
h264-x264.h:379: error: braces around scalar initializer for type 'signed 
char:1'
h264-x264.h:390: error: braces around scalar initializer for type 'signed 
char:1'
h264-x264.h:390: error: braces around scalar initializer for type 'signed 
char:1'
h264-x264.h:402: error: braces around scalar initializer for type 'signed 
char:1'
h264-x264.h:402: error: braces around scalar initializer for type 'signed 
char:1'
h264-x264.h:414: error: braces around scalar initializer for type 'signed 
char:1'
h264-x264.h:414: error: braces around scalar initializer for type 'signed 
char:1'
h264-x264.h:426: error: braces around scalar initializer for type 'signed 
char:1'
h264-x264.h:426: error: braces around scalar initializer for type 'signed 
char:1'

the problem is the macro DECLARE_GENERIC_OPTIONS(...) in h264-x264.h
and a patch is attached, which let it compile;

but: in Ekiga it shows H.264 again only for SIP and not for H.323; do I
have to activate something else?

Emilio
-- 
Matthias Apitz
*** opal/plugins/video/H.264/h264-x264.h	2008-04-02 17:07:24.0 +0200
--- opal/plugins/video/H.264/h264-x264.h.orig	2008-04-02 17:06:02.0 +0200
***
*** 275,282 
  }; \
  static const struct PluginCodec_H323GenericParameterDefinition prefix##_h323params[] = \
  { \
! 	{1,41, PluginCodec_H323GenericParameterDefinition::PluginCodec_GenericParameter_BooleanArray, prefix##_Profile }, \
! 	{1,42, PluginCodec_H323GenericParameterDefinition::PluginCodec_GenericParameter_unsignedMin, prefix##_Level }, \
  	NULL \
  }; \
  static struct PluginCodec_H323GenericCodecData prefix##_h323GenericData[] = { \
--- 275,282 
  }; \
  static const struct PluginCodec_H323GenericParameterDefinition prefix##_h323params[] = \
  { \
! 	{1,41, PluginCodec_H323GenericParameterDefinition::PluginCodec_GenericParameter_BooleanArray,{prefix##_Profile}}, \
! 	{1,42, PluginCodec_H323GenericParameterDefinition::PluginCodec_GenericParameter_unsignedMin,{prefix##_Level}}, \
  	NULL \
  }; \
  static struct PluginCodec_H323GenericCodecData prefix##_h323GenericData[] = { \
___
Ekiga-devel-list mailing list
Ekiga-devel-list@gnome.org
http://mail.gnome.org/mailman/listinfo/ekiga-devel-list

Re: [Ekiga-devel-list] 1st working SVN build Ekiga on FreeBSD

2008-04-02 Thread Matthias Apitz

El día Wednesday, April 02, 2008 a las 03:29:30PM +0200, Matthias Schneider 
escribió:

> > I don't even know what I should see as echo from [EMAIL PROTECTED], the echo
> > of my own cam picture like looking into a mirror or something else?
> The image you are transmitting is looped back, so if you transmit a moving 
> logo
> you get the moving logo, if you transmit a camera signal you get the camera
> signal.

ok;

> > > i.e. is the on_established_cb still
> > > not being called? Do I understand correctly it does/did not work with V4L,
> > but
> > > with the moving logo it worked?
> >
> > yes, and yes; with V4L aktivated the call does not reach
> > OnEstablished(), can't be hanged'up but the voice is echoed correctly;
> > with moving logo all works fine;
> >
> > I have had no further time until now to debug; do you have an idea or do
> > you working on this as well?
> 
> Hm, it seems you never get a single video RTP packet. Could you verify that 
> with
> a protocol analyzer? Another thing possible would be if the receiving video
> thread crashes at some point. Does preview work with V4L?

I did H.323 tests against someone using Netmeeting (which worked fine)
and against our VSX7000 VC system (in which at least I can see what the
VSX7000 is sending as video); both tests have be done inside our VPN;
I'll check with tcpdump if in connection to [EMAIL PROTECTED] via Internet
it is not our firewall that disables the receiving...

> > do you have any plans about H.264 in H.323?
> None from my side but there is an experimental implmenetation by Simone Horne
> you can activate via a simple #define.
> Matthias

in which part is this, or what is the #define?

thx

Emilio

-- 
Matthias Apitz
Manager Technical Support - OCLC GmbH
Gruenwalder Weg 28g - 82041 Oberhaching - Germany
t +49-89-61308 351 - f +49-89-61308 399 - m +49-170-4527211
e <[EMAIL PROTECTED]> - w http://www.oclc.org/ http://www.UnixArea.de/
b http://gurucubano.blogspot.com/
Don't top-post, read RFC1855 http://www.faqs.org/rfcs/rfc1855.html
A: Because it messes up the order in which people normally read text.
Q: Why is it such a bad thing?
A: Top-posting.
Q: What is the most annoying thing on Usenet and in e-mail?
___
Ekiga-devel-list mailing list
Ekiga-devel-list@gnome.org
http://mail.gnome.org/mailman/listinfo/ekiga-devel-list

Re: [Ekiga-devel-list] 1st working SVN build Ekiga on FreeBSD

2008-04-02 Thread Matthias Schneider

Quoting Matthias Apitz <[EMAIL PROTECTED]>:

> El día Wednesday, April 02, 2008 a las 12:18:25PM +0200, Matthias Schneider
> escribió:
>
> > Quoting Matthias Apitz <[EMAIL PROTECTED]>:
> >
> > > El día Friday, March 28, 2008 a las 12:16:53PM +0100, Damien Sandras
> > > escribió:
> > >
> > > > Try changing the video size ?
> > > >
> > > > I think you will have to run things step by step, it will be boring,
> but
> > > > I'm sure it will fix a severe bug that other persons could encounter.
> > >
> > > We're getting closer... I've put in some debug fprintf's and for
> > > [EMAIL PROTECTED] it asks OPAL to do:
> > >
> > > Debug:lib/engine/protocol/skel/call-core.cpp CallCore::dial()
> > > Debug:src/endpoints/opal-call.cpp Opal::Call::Call()
> > > Debug:src/endpoints/opal-call.cpp Opal::Call::OnOpenMediaStream()
> > > Debug:StreamType Audio
> > > Debug:stream_name: PCMU
> > > Debug:src/endpoints/opal-call.cpp Opal::Call::OnOpenMediaStream()
> > > Debug:StreamType Audio
> > > Debug:stream_name: PCMU
> > > Debug:src/endpoints/opal-call.cpp Opal::Call::OnEstablished()
> > > Debug:src/endpoints/opal-call.cpp Opal::Call::OnEstablished()
> > > Debug:on_established_call_cb
> > > Debug:setting mw->current_call = &call;
> > >
> > > i.e. no video is involved (which explains as well that switching off V4L
> > > to Moving Logo works well);
> > >
> > > while for [EMAIL PROTECTED] it is:
> > >
> > > Debug:lib/engine/protocol/skel/call-core.cpp CallCore::dial()
> > > Debug:src/endpoints/opal-call.cpp Opal::Call::Call()
> > > Debug:lib/engine/protocol/skel/call-core.cpp CallCore::dial()
> > > Debug:src/endpoints/opal-call.cpp Opal::Call::Call()
> > > Debug:src/endpoints/opal-call.cpp Opal::Call::OnOpenMediaStream()
> > > Debug:StreamType Audio
> > > Debug:stream_name: PCMU
> > > Debug:src/endpoints/opal-call.cpp Opal::Call::OnOpenMediaStream()
> > > Debug:StreamType Audio
> > > Debug:stream_name: PCMU
> > > Debug:src/endpoints/opal-call.cpp Opal::Call::OnOpenMediaStream()
> > > Debug:StreamType Video
> > > Debug:stream_name: h261
> > > Debug:src/endpoints/opal-call.cpp Opal::Call::OnOpenMediaStream()
> > > Debug:StreamType Video
> > > Debug:stream_name: h261
> > >
> > > and OnEstablished() is not passed...
> > >
> > > does [EMAIL PROTECTED] echoes video?
> > >
> > > changing the video size to 640x480 has no effect on this;
> > >
> > >   matthias
> > >
> > > --
> > > Matthias Apitz
> >
> >
> > Matthias,
>
> Hi Matthias,
>
> I think we have now the problem here that we have (at least) two
> 'Matthias', even with the same spelling; I propose that I firm from now
> on and should be called 'Emilio' (which is my 2nd name in German form
> as 'Emil');
>
> >
> > Damien has commited a patch to always include libgen.h, which is also a
> good
> > idea on linux.
> >
> > May I ask if you still do not get any video,
>
> I don't even know what I should see as echo from [EMAIL PROTECTED], the echo
> of my own cam picture like looking into a mirror or something else?
The image you are transmitting is looped back, so if you transmit a moving logo
you get the moving logo, if you transmit a camera signal you get the camera
signal.

> > i.e. is the on_established_cb still
> > not being called? Do I understand correctly it does/did not work with V4L,
> but
> > with the moving logo it worked?
>
> yes, and yes; with V4L aktivated the call does not reach
> OnEstablished(), can't be hanged'up but the voice is echoed correctly;
> with moving logo all works fine;
>
> I have had no further time until now to debug; do you have an idea or do
> you working on this as well?

Hm, it seems you never get a single video RTP packet. Could you verify that with
a protocol analyzer? Another thing possible would be if the receiving video
thread crashes at some point. Does preview work with V4L?

> do you have any plans about H.264 in H.323?
None from my side but there is an experimental implmenetation by Simone Horne
you can activate via a simple #define.
Matthias

>   Emilio
> --
> Matthias Emilio Apitz
> b http://gurucubano.blogspot.com/
> Don't top-post, read RFC1855 http://www.faqs.org/rfcs/rfc1855.html
> A: Because it messes up the order in which people normally read text.
> Q: Why is it such a bad thing?
> A: Top-posting.
> Q: What is the most annoying thing on Usenet and in e-mail?
>





This message was sent using IMP, the Internet Messaging Program.
___
Ekiga-devel-list mailing list
Ekiga-devel-list@gnome.org
http://mail.gnome.org/mailman/listinfo/ekiga-devel-list

Re: [Ekiga-devel-list] 1st working SVN build Ekiga on FreeBSD

2008-04-02 Thread Matthias Apitz

El día Wednesday, April 02, 2008 a las 12:18:25PM +0200, Matthias Schneider 
escribió:

> Quoting Matthias Apitz <[EMAIL PROTECTED]>:
> 
> > El día Friday, March 28, 2008 a las 12:16:53PM +0100, Damien Sandras
> > escribió:
> >
> > > Try changing the video size ?
> > >
> > > I think you will have to run things step by step, it will be boring, but
> > > I'm sure it will fix a severe bug that other persons could encounter.
> >
> > We're getting closer... I've put in some debug fprintf's and for
> > [EMAIL PROTECTED] it asks OPAL to do:
> >
> > Debug:lib/engine/protocol/skel/call-core.cpp CallCore::dial()
> > Debug:src/endpoints/opal-call.cpp Opal::Call::Call()
> > Debug:src/endpoints/opal-call.cpp Opal::Call::OnOpenMediaStream()
> > Debug:StreamType Audio
> > Debug:stream_name: PCMU
> > Debug:src/endpoints/opal-call.cpp Opal::Call::OnOpenMediaStream()
> > Debug:StreamType Audio
> > Debug:stream_name: PCMU
> > Debug:src/endpoints/opal-call.cpp Opal::Call::OnEstablished()
> > Debug:src/endpoints/opal-call.cpp Opal::Call::OnEstablished()
> > Debug:on_established_call_cb
> > Debug:setting mw->current_call = &call;
> >
> > i.e. no video is involved (which explains as well that switching off V4L
> > to Moving Logo works well);
> >
> > while for [EMAIL PROTECTED] it is:
> >
> > Debug:lib/engine/protocol/skel/call-core.cpp CallCore::dial()
> > Debug:src/endpoints/opal-call.cpp Opal::Call::Call()
> > Debug:lib/engine/protocol/skel/call-core.cpp CallCore::dial()
> > Debug:src/endpoints/opal-call.cpp Opal::Call::Call()
> > Debug:src/endpoints/opal-call.cpp Opal::Call::OnOpenMediaStream()
> > Debug:StreamType Audio
> > Debug:stream_name: PCMU
> > Debug:src/endpoints/opal-call.cpp Opal::Call::OnOpenMediaStream()
> > Debug:StreamType Audio
> > Debug:stream_name: PCMU
> > Debug:src/endpoints/opal-call.cpp Opal::Call::OnOpenMediaStream()
> > Debug:StreamType Video
> > Debug:stream_name: h261
> > Debug:src/endpoints/opal-call.cpp Opal::Call::OnOpenMediaStream()
> > Debug:StreamType Video
> > Debug:stream_name: h261
> >
> > and OnEstablished() is not passed...
> >
> > does [EMAIL PROTECTED] echoes video?
> >
> > changing the video size to 640x480 has no effect on this;
> >
> > matthias
> >
> > --
> > Matthias Apitz
> 
> 
> Matthias,

Hi Matthias,

I think we have now the problem here that we have (at least) two
'Matthias', even with the same spelling; I propose that I firm from now
on and should be called 'Emilio' (which is my 2nd name in German form
as 'Emil');

> 
> Damien has commited a patch to always include libgen.h, which is also a good
> idea on linux.
> 
> May I ask if you still do not get any video,

I don't even know what I should see as echo from [EMAIL PROTECTED], the echo
of my own cam picture like looking into a mirror or something else?

> i.e. is the on_established_cb still
> not being called? Do I understand correctly it does/did not work with V4L, but
> with the moving logo it worked?

yes, and yes; with V4L aktivated the call does not reach
OnEstablished(), can't be hanged'up but the voice is echoed correctly;
with moving logo all works fine;

I have had no further time until now to debug; do you have an idea or do
you working on this as well?

do you have any plans about H.264 in H.323?

Emilio
-- 
Matthias Emilio Apitz
b http://gurucubano.blogspot.com/
Don't top-post, read RFC1855 http://www.faqs.org/rfcs/rfc1855.html
A: Because it messes up the order in which people normally read text.
Q: Why is it such a bad thing?
A: Top-posting.
Q: What is the most annoying thing on Usenet and in e-mail?
___
Ekiga-devel-list mailing list
Ekiga-devel-list@gnome.org
http://mail.gnome.org/mailman/listinfo/ekiga-devel-list

Re: [Ekiga-devel-list] 1st working SVN build Ekiga on FreeBSD

2008-04-02 Thread Matthias Schneider

Quoting Matthias Apitz <[EMAIL PROTECTED]>:

> El día Friday, March 28, 2008 a las 12:16:53PM +0100, Damien Sandras
> escribió:
>
> > Try changing the video size ?
> >
> > I think you will have to run things step by step, it will be boring, but
> > I'm sure it will fix a severe bug that other persons could encounter.
>
> We're getting closer... I've put in some debug fprintf's and for
> [EMAIL PROTECTED] it asks OPAL to do:
>
> Debug:lib/engine/protocol/skel/call-core.cpp CallCore::dial()
> Debug:src/endpoints/opal-call.cpp Opal::Call::Call()
> Debug:src/endpoints/opal-call.cpp Opal::Call::OnOpenMediaStream()
> Debug:StreamType Audio
> Debug:stream_name: PCMU
> Debug:src/endpoints/opal-call.cpp Opal::Call::OnOpenMediaStream()
> Debug:StreamType Audio
> Debug:stream_name: PCMU
> Debug:src/endpoints/opal-call.cpp Opal::Call::OnEstablished()
> Debug:src/endpoints/opal-call.cpp Opal::Call::OnEstablished()
> Debug:on_established_call_cb
> Debug:setting mw->current_call = &call;
>
> i.e. no video is involved (which explains as well that switching off V4L
> to Moving Logo works well);
>
> while for [EMAIL PROTECTED] it is:
>
> Debug:lib/engine/protocol/skel/call-core.cpp CallCore::dial()
> Debug:src/endpoints/opal-call.cpp Opal::Call::Call()
> Debug:lib/engine/protocol/skel/call-core.cpp CallCore::dial()
> Debug:src/endpoints/opal-call.cpp Opal::Call::Call()
> Debug:src/endpoints/opal-call.cpp Opal::Call::OnOpenMediaStream()
> Debug:StreamType Audio
> Debug:stream_name: PCMU
> Debug:src/endpoints/opal-call.cpp Opal::Call::OnOpenMediaStream()
> Debug:StreamType Audio
> Debug:stream_name: PCMU
> Debug:src/endpoints/opal-call.cpp Opal::Call::OnOpenMediaStream()
> Debug:StreamType Video
> Debug:stream_name: h261
> Debug:src/endpoints/opal-call.cpp Opal::Call::OnOpenMediaStream()
> Debug:StreamType Video
> Debug:stream_name: h261
>
> and OnEstablished() is not passed...
>
> does [EMAIL PROTECTED] echoes video?
>
> changing the video size to 640x480 has no effect on this;
>
>   matthias
>
> --
> Matthias Apitz


Matthias,

Damien has commited a patch to always include libgen.h, which is also a good
idea on linux.

May I ask if you still do not get any video, i.e. is the on_established_cb still
not being called? Do I understand correctly it does/did not work with V4L, but
with the moving logo it worked?

Matthias





This message was sent using IMP, the Internet Messaging Program.
___
Ekiga-devel-list mailing list
Ekiga-devel-list@gnome.org
http://mail.gnome.org/mailman/listinfo/ekiga-devel-list

Re: [Ekiga-devel-list] 1st working SVN build Ekiga on FreeBSD

2008-03-28 Thread Matthias Apitz

El día Friday, March 28, 2008 a las 12:16:53PM +0100, Damien Sandras escribió:

> Try changing the video size ?
> 
> I think you will have to run things step by step, it will be boring, but
> I'm sure it will fix a severe bug that other persons could encounter.

We're getting closer... I've put in some debug fprintf's and for
[EMAIL PROTECTED] it asks OPAL to do:

Debug:lib/engine/protocol/skel/call-core.cpp CallCore::dial()
Debug:src/endpoints/opal-call.cpp Opal::Call::Call()
Debug:src/endpoints/opal-call.cpp Opal::Call::OnOpenMediaStream()
Debug:StreamType Audio
Debug:stream_name: PCMU
Debug:src/endpoints/opal-call.cpp Opal::Call::OnOpenMediaStream()
Debug:StreamType Audio
Debug:stream_name: PCMU
Debug:src/endpoints/opal-call.cpp Opal::Call::OnEstablished()
Debug:src/endpoints/opal-call.cpp Opal::Call::OnEstablished()
Debug:on_established_call_cb
Debug:setting mw->current_call = &call;

i.e. no video is involved (which explains as well that switching off V4L
to Moving Logo works well);

while for [EMAIL PROTECTED] it is:

Debug:lib/engine/protocol/skel/call-core.cpp CallCore::dial()
Debug:src/endpoints/opal-call.cpp Opal::Call::Call()
Debug:lib/engine/protocol/skel/call-core.cpp CallCore::dial()
Debug:src/endpoints/opal-call.cpp Opal::Call::Call()
Debug:src/endpoints/opal-call.cpp Opal::Call::OnOpenMediaStream()
Debug:StreamType Audio
Debug:stream_name: PCMU
Debug:src/endpoints/opal-call.cpp Opal::Call::OnOpenMediaStream()
Debug:StreamType Audio
Debug:stream_name: PCMU
Debug:src/endpoints/opal-call.cpp Opal::Call::OnOpenMediaStream()
Debug:StreamType Video
Debug:stream_name: h261
Debug:src/endpoints/opal-call.cpp Opal::Call::OnOpenMediaStream()
Debug:StreamType Video
Debug:stream_name: h261

and OnEstablished() is not passed...

does [EMAIL PROTECTED] echoes video? 

changing the video size to 640x480 has no effect on this;

matthias

-- 
Matthias Apitz
Manager Technical Support - OCLC GmbH
Gruenwalder Weg 28g - 82041 Oberhaching - Germany
t +49-89-61308 351 - f +49-89-61308 399 - m +49-170-4527211
e <[EMAIL PROTECTED]> - w http://www.oclc.org/ http://www.UnixArea.de/
b http://gurucubano.blogspot.com/
Don't top-post, read RFC1855 http://www.faqs.org/rfcs/rfc1855.html
A: Because it messes up the order in which people normally read text.
Q: Why is it such a bad thing?
A: Top-posting.
Q: What is the most annoying thing on Usenet and in e-mail?
___
Ekiga-devel-list mailing list
Ekiga-devel-list@gnome.org
http://mail.gnome.org/mailman/listinfo/ekiga-devel-list

Re: [Ekiga-devel-list] 1st working SVN build Ekiga on FreeBSD

2008-03-28 Thread Damien Sandras

Le vendredi 28 mars 2008 à 11:37 +0100, Matthias Apitz a écrit :
> El día Friday, March 28, 2008 a las 10:19:05AM +0100, Damien Sandras escribió:
> 
> > Le vendredi 28 mars 2008 à 07:13 +0100, Matthias Apitz a écrit :
> > > El día Thursday, March 27, 2008 a las 08:10:28PM +0100, Damien Sandras 
> > > escribió:
> > > 
> > > > > > That's because the call never reaches the established state...
> > > > > 
> > > > > Yes, this was my conclusion as well that on_established_call_cb() is 
> > > > > not
> > > > > get called, even if voice echo to/from [EMAIL PROTECTED] works fine;
> > > > > > 
> > > > > > Can you try with and without video ?
> > > > > 
> > > > > when I set video from V4L to MovingLogo it works,
> > > > > on_established_call_cb() gets called and hang-up is ok;
> > > > > 
> > > > 
> > > > Can you post a -d 4 somewhere ?
> > > > 
> > > > But I have the feeling the video driver does not allow some method to
> > > > return, preventing the call flow to behave normally.
> > > 
> > > Here you go: http://www.unixarea.de/Ekiga/ekigaLogD4.txt
> > 
> > Unfortunately, I see nothing wrong there...
> > 
> > Can you try with other resolutions, less codecs, ... and try to isolate
> > the problem ?
> 
> Have reduced the codecs to only H.261 for video and PCMU for audio, same
> result; will do further debugging on the weekend;
> 

Try changing the video size ?

I think you will have to run things step by step, it will be boring, but
I'm sure it will fix a severe bug that other persons could encounter.

Thanks for your time !
-- 
 _ Damien Sandras
(o-  
//\Ekiga Softphone : http://www.ekiga.org/
v_/_   NOVACOM : http://www.novacom.be/
   FOSDEM  : http://www.fosdem.org/
   SIP Phone   : sip:[EMAIL PROTECTED]
   

___
Ekiga-devel-list mailing list
Ekiga-devel-list@gnome.org
http://mail.gnome.org/mailman/listinfo/ekiga-devel-list

Re: [Ekiga-devel-list] 1st working SVN build Ekiga on FreeBSD

2008-03-28 Thread Matthias Apitz

El día Friday, March 28, 2008 a las 10:19:05AM +0100, Damien Sandras escribió:

> Le vendredi 28 mars 2008 à 07:13 +0100, Matthias Apitz a écrit :
> > El día Thursday, March 27, 2008 a las 08:10:28PM +0100, Damien Sandras 
> > escribió:
> > 
> > > > > That's because the call never reaches the established state...
> > > > 
> > > > Yes, this was my conclusion as well that on_established_call_cb() is not
> > > > get called, even if voice echo to/from [EMAIL PROTECTED] works fine;
> > > > > 
> > > > > Can you try with and without video ?
> > > > 
> > > > when I set video from V4L to MovingLogo it works,
> > > > on_established_call_cb() gets called and hang-up is ok;
> > > > 
> > > 
> > > Can you post a -d 4 somewhere ?
> > > 
> > > But I have the feeling the video driver does not allow some method to
> > > return, preventing the call flow to behave normally.
> > 
> > Here you go: http://www.unixarea.de/Ekiga/ekigaLogD4.txt
> 
> Unfortunately, I see nothing wrong there...
> 
> Can you try with other resolutions, less codecs, ... and try to isolate
> the problem ?

Have reduced the codecs to only H.261 for video and PCMU for audio, same
result; will do further debugging on the weekend;

matthias

-- 
Matthias Apitz
Manager Technical Support - OCLC GmbH
Gruenwalder Weg 28g - 82041 Oberhaching - Germany
t +49-89-61308 351 - f +49-89-61308 399 - m +49-170-4527211
e <[EMAIL PROTECTED]> - w http://www.oclc.org/ http://www.UnixArea.de/
b http://gurucubano.blogspot.com/
Don't top-post, read RFC1855 http://www.faqs.org/rfcs/rfc1855.html
A: Because it messes up the order in which people normally read text.
Q: Why is it such a bad thing?
A: Top-posting.
Q: What is the most annoying thing on Usenet and in e-mail?
___
Ekiga-devel-list mailing list
Ekiga-devel-list@gnome.org
http://mail.gnome.org/mailman/listinfo/ekiga-devel-list

Re: [Ekiga-devel-list] 1st working SVN build Ekiga on FreeBSD

2008-03-28 Thread Damien Sandras

Le vendredi 28 mars 2008 à 07:13 +0100, Matthias Apitz a écrit :
> El día Thursday, March 27, 2008 a las 08:10:28PM +0100, Damien Sandras 
> escribió:
> 
> > > > That's because the call never reaches the established state...
> > > 
> > > Yes, this was my conclusion as well that on_established_call_cb() is not
> > > get called, even if voice echo to/from [EMAIL PROTECTED] works fine;
> > > > 
> > > > Can you try with and without video ?
> > > 
> > > when I set video from V4L to MovingLogo it works,
> > > on_established_call_cb() gets called and hang-up is ok;
> > > 
> > 
> > Can you post a -d 4 somewhere ?
> > 
> > But I have the feeling the video driver does not allow some method to
> > return, preventing the call flow to behave normally.
> 
> Here you go: http://www.unixarea.de/Ekiga/ekigaLogD4.txt

Unfortunately, I see nothing wrong there...

Can you try with other resolutions, less codecs, ... and try to isolate
the problem ?

I have discovered a problem in the log, but I have the same problem, so
I do not think it causes the behavior you describe. I will fix it this
week-end.
-- 
 _ Damien Sandras
(o-  
//\Ekiga Softphone : http://www.ekiga.org/
v_/_   NOVACOM : http://www.novacom.be/
   FOSDEM  : http://www.fosdem.org/
   SIP Phone   : sip:[EMAIL PROTECTED]
   

___
Ekiga-devel-list mailing list
Ekiga-devel-list@gnome.org
http://mail.gnome.org/mailman/listinfo/ekiga-devel-list

Re: [Ekiga-devel-list] 1st working SVN build Ekiga on FreeBSD

2008-03-28 Thread Matthias Apitz

El día Thursday, March 27, 2008 a las 08:10:28PM +0100, Damien Sandras escribió:

> > > That's because the call never reaches the established state...
> > 
> > Yes, this was my conclusion as well that on_established_call_cb() is not
> > get called, even if voice echo to/from [EMAIL PROTECTED] works fine;
> > > 
> > > Can you try with and without video ?
> > 
> > when I set video from V4L to MovingLogo it works,
> > on_established_call_cb() gets called and hang-up is ok;
> > 
> 
> Can you post a -d 4 somewhere ?
> 
> But I have the feeling the video driver does not allow some method to
> return, preventing the call flow to behave normally.

Here you go: http://www.unixarea.de/Ekiga/ekigaLogD4.txt

matthias
-- 
Matthias Apitz
Manager Technical Support - OCLC GmbH
Gruenwalder Weg 28g - 82041 Oberhaching - Germany
t +49-89-61308 351 - f +49-89-61308 399 - m +49-170-4527211
e <[EMAIL PROTECTED]> - w http://www.oclc.org/ http://www.UnixArea.de/
b http://gurucubano.blogspot.com/
Don't top-post, read RFC1855 http://www.faqs.org/rfcs/rfc1855.html
A: Because it messes up the order in which people normally read text.
Q: Why is it such a bad thing?
A: Top-posting.
Q: What is the most annoying thing on Usenet and in e-mail?
___
Ekiga-devel-list mailing list
Ekiga-devel-list@gnome.org
http://mail.gnome.org/mailman/listinfo/ekiga-devel-list

Re: [Ekiga-devel-list] 1st working SVN build Ekiga on FreeBSD

2008-03-27 Thread Damien Sandras

Le jeudi 27 mars 2008 à 19:54 +0100, Matthias Apitz a écrit :
> El día Thursday, March 27, 2008 a las 05:38:38PM +0100, Damien Sandras 
> escribió:
> 
> > > toggle_call_cb (GtkWidget * widget,
> > > gpointer data)
> > > {
> > >   GmMainWindow *mw = NULL;
> > > 
> > >   std::cout << "Debug:toggle_call_cb()" << std::endl << std::flush;
> > >   mw = gm_mw_get_mw (GTK_WIDGET (data));
> > > 
> > >   if (mw->current_call == NULL)
> > > std::cout << "Debug: no mw->current_call pointer" << std::endl << 
> > > std::flush;
> > >   if (gm_connect_button_get_connected (GM_CONNECT_BUTTON 
> > > (mw->connect_button)))
> > > std::cout << "gm_connect_button_get_connected... is true" << 
> > > std::endl << std::flush;
> > > 
> > >   if (mw->current_call && gm_connect_button_get_connected 
> > > (GM_CONNECT_BUTTON (mw->connect_button))) {
> > > std::cout << "Debug:calling hangup_call_cb now" << std::endl << 
> > > std::flush;
> > > hangup_call_cb (widget, data);
> > >   }
> > >   else if (!mw->current_call && !gm_connect_button_get_connected 
> > > (GM_CONNECT_BUTTON (mw->connect_button))) {
> > > std::cout << "Debug:calling place_call_cb now" << std::endl << 
> > > std::flush;
> > > place_call_cb (widget, data);
> > >   }
> > >   std::cout << "Debug:did nothing in toggle_call_cb" << std::endl << 
> > > std::flush;
> > > }
> > > 
> > > in a call to [EMAIL PROTECTED] mw->current_call is NULL and nothing is 
> > > done
> > > in this function;
> > > 
> > 
> > That's because the call never reaches the established state...
> 
> Yes, this was my conclusion as well that on_established_call_cb() is not
> get called, even if voice echo to/from [EMAIL PROTECTED] works fine;
> > 
> > Can you try with and without video ?
> 
> when I set video from V4L to MovingLogo it works,
> on_established_call_cb() gets called and hang-up is ok;
> 

Can you post a -d 4 somewhere ?

But I have the feeling the video driver does not allow some method to
return, preventing the call flow to behave normally.
-- 
 _ Damien Sandras
(o-  
//\Ekiga Softphone : http://www.ekiga.org/
v_/_   NOVACOM : http://www.novacom.be/
   FOSDEM  : http://www.fosdem.org/
   SIP Phone   : sip:[EMAIL PROTECTED]
   

___
Ekiga-devel-list mailing list
Ekiga-devel-list@gnome.org
http://mail.gnome.org/mailman/listinfo/ekiga-devel-list

Re: [Ekiga-devel-list] 1st working SVN build Ekiga on FreeBSD

2008-03-27 Thread Matthias Apitz

El día Thursday, March 27, 2008 a las 05:38:38PM +0100, Damien Sandras escribió:

> > toggle_call_cb (GtkWidget * widget,
> > gpointer data)
> > {
> >   GmMainWindow *mw = NULL;
> > 
> >   std::cout << "Debug:toggle_call_cb()" << std::endl << std::flush;
> >   mw = gm_mw_get_mw (GTK_WIDGET (data));
> > 
> >   if (mw->current_call == NULL)
> > std::cout << "Debug: no mw->current_call pointer" << std::endl << 
> > std::flush;
> >   if (gm_connect_button_get_connected (GM_CONNECT_BUTTON 
> > (mw->connect_button)))
> > std::cout << "gm_connect_button_get_connected... is true" << std::endl 
> > << std::flush;
> > 
> >   if (mw->current_call && gm_connect_button_get_connected 
> > (GM_CONNECT_BUTTON (mw->connect_button))) {
> > std::cout << "Debug:calling hangup_call_cb now" << std::endl << 
> > std::flush;
> > hangup_call_cb (widget, data);
> >   }
> >   else if (!mw->current_call && !gm_connect_button_get_connected 
> > (GM_CONNECT_BUTTON (mw->connect_button))) {
> > std::cout << "Debug:calling place_call_cb now" << std::endl << 
> > std::flush;
> > place_call_cb (widget, data);
> >   }
> >   std::cout << "Debug:did nothing in toggle_call_cb" << std::endl << 
> > std::flush;
> > }
> > 
> > in a call to [EMAIL PROTECTED] mw->current_call is NULL and nothing is done
> > in this function;
> > 
> 
> That's because the call never reaches the established state...

Yes, this was my conclusion as well that on_established_call_cb() is not
get called, even if voice echo to/from [EMAIL PROTECTED] works fine;
> 
> Can you try with and without video ?

when I set video from V4L to MovingLogo it works,
on_established_call_cb() gets called and hang-up is ok;

matthias
-- 
Matthias Apitz
Manager Technical Support - OCLC GmbH
Gruenwalder Weg 28g - 82041 Oberhaching - Germany
t +49-89-61308 351 - f +49-89-61308 399 - m +49-170-4527211
e <[EMAIL PROTECTED]> - w http://www.oclc.org/ http://www.UnixArea.de/
b http://gurucubano.blogspot.com/
Don't top-post, read RFC1855 http://www.faqs.org/rfcs/rfc1855.html
A: Because it messes up the order in which people normally read text.
Q: Why is it such a bad thing?
A: Top-posting.
Q: What is the most annoying thing on Usenet and in e-mail?
___
Ekiga-devel-list mailing list
Ekiga-devel-list@gnome.org
http://mail.gnome.org/mailman/listinfo/ekiga-devel-list

Re: [Ekiga-devel-list] 1st working SVN build Ekiga on FreeBSD

2008-03-27 Thread Damien Sandras

Le jeudi 27 mars 2008 à 17:14 +0100, Matthias Apitz a écrit :
> El día Thursday, March 27, 2008 a las 01:54:44PM +0100, Damien Sandras 
> escribió:
> 
> > > I disagree :-)
> > > 
> > > in the call to [EMAIL PROTECTED], which can't be hang-up, the OSS
> > > PSoundChannelOSS::Close() is not called, while it is in the call with
> > > [EMAIL PROTECTED]; I proofed it with a debug statement in
> > > plugins/sound_oss/sound_oss.cxx which prints:
> > > 2008/03/27 13:36:36.038   0:37.168 OnRelease:0x2be16e00 OSS-debug 
> > >   PSoundChannelOSS::Close() called
> > > 2008/03/27 13:36:36.045   0:37.175 OnRelease:0x2be16e00 OSS-debug 
> > >   PSoundChannelOSS::Close() device /dev/dsp0
> > > but only in the *010600 call; the problem must be in higher level;
> > > 
> > 
> > Can you see where it deadlocks in that case ?
> > 
> > We have no special code for [EMAIL PROTECTED] which could explain a
> > different behavior for that specific url :-)
> 
> in src/gui/main.cpp in the function:
> 
> static void
> toggle_call_cb (GtkWidget * widget,
> gpointer data)
> {
>   GmMainWindow *mw = NULL;
> 
>   std::cout << "Debug:toggle_call_cb()" << std::endl << std::flush;
>   mw = gm_mw_get_mw (GTK_WIDGET (data));
> 
>   if (mw->current_call == NULL)
> std::cout << "Debug: no mw->current_call pointer" << std::endl << 
> std::flush;
>   if (gm_connect_button_get_connected (GM_CONNECT_BUTTON 
> (mw->connect_button)))
> std::cout << "gm_connect_button_get_connected... is true" << std::endl << 
> std::flush;
> 
>   if (mw->current_call && gm_connect_button_get_connected (GM_CONNECT_BUTTON 
> (mw->connect_button))) {
> std::cout << "Debug:calling hangup_call_cb now" << std::endl << 
> std::flush;
> hangup_call_cb (widget, data);
>   }
>   else if (!mw->current_call && !gm_connect_button_get_connected 
> (GM_CONNECT_BUTTON (mw->connect_button))) {
> std::cout << "Debug:calling place_call_cb now" << std::endl << std::flush;
> place_call_cb (widget, data);
>   }
>   std::cout << "Debug:did nothing in toggle_call_cb" << std::endl << 
> std::flush;
> }
> 
> in a call to [EMAIL PROTECTED] mw->current_call is NULL and nothing is done
> in this function;
> 

That's because the call never reaches the established state...

Can you try with and without video ?
-- 
 _ Damien Sandras
(o-  
//\Ekiga Softphone : http://www.ekiga.org/
v_/_   NOVACOM : http://www.novacom.be/
   FOSDEM  : http://www.fosdem.org/
   SIP Phone   : sip:[EMAIL PROTECTED]
   

___
Ekiga-devel-list mailing list
Ekiga-devel-list@gnome.org
http://mail.gnome.org/mailman/listinfo/ekiga-devel-list

Re: [Ekiga-devel-list] 1st working SVN build Ekiga on FreeBSD

2008-03-27 Thread Matthias Apitz

El día Thursday, March 27, 2008 a las 01:54:44PM +0100, Damien Sandras escribió:

> > I disagree :-)
> > 
> > in the call to [EMAIL PROTECTED], which can't be hang-up, the OSS
> > PSoundChannelOSS::Close() is not called, while it is in the call with
> > [EMAIL PROTECTED]; I proofed it with a debug statement in
> > plugins/sound_oss/sound_oss.cxx which prints:
> > 2008/03/27 13:36:36.038   0:37.168 OnRelease:0x2be16e00 OSS-debug   
> > PSoundChannelOSS::Close() called
> > 2008/03/27 13:36:36.045   0:37.175 OnRelease:0x2be16e00 OSS-debug   
> > PSoundChannelOSS::Close() device /dev/dsp0
> > but only in the *010600 call; the problem must be in higher level;
> > 
> 
> Can you see where it deadlocks in that case ?
> 
> We have no special code for [EMAIL PROTECTED] which could explain a
> different behavior for that specific url :-)

in src/gui/main.cpp in the function:

static void
toggle_call_cb (GtkWidget * widget,
gpointer data)
{
  GmMainWindow *mw = NULL;

  std::cout << "Debug:toggle_call_cb()" << std::endl << std::flush;
  mw = gm_mw_get_mw (GTK_WIDGET (data));

  if (mw->current_call == NULL)
std::cout << "Debug: no mw->current_call pointer" << std::endl << 
std::flush;
  if (gm_connect_button_get_connected (GM_CONNECT_BUTTON (mw->connect_button)))
std::cout << "gm_connect_button_get_connected... is true" << std::endl << 
std::flush;

  if (mw->current_call && gm_connect_button_get_connected (GM_CONNECT_BUTTON 
(mw->connect_button))) {
std::cout << "Debug:calling hangup_call_cb now" << std::endl << std::flush;
hangup_call_cb (widget, data);
  }
  else if (!mw->current_call && !gm_connect_button_get_connected 
(GM_CONNECT_BUTTON (mw->connect_button))) {
std::cout << "Debug:calling place_call_cb now" << std::endl << std::flush;
place_call_cb (widget, data);
  }
  std::cout << "Debug:did nothing in toggle_call_cb" << std::endl << std::flush;
}

in a call to [EMAIL PROTECTED] mw->current_call is NULL and nothing is done
in this function;

matthias
-- 
Matthias Apitz
Manager Technical Support - OCLC GmbH
Gruenwalder Weg 28g - 82041 Oberhaching - Germany
t +49-89-61308 351 - f +49-89-61308 399 - m +49-170-4527211
e <[EMAIL PROTECTED]> - w http://www.oclc.org/ http://www.UnixArea.de/
b http://gurucubano.blogspot.com/
Don't top-post, read RFC1855 http://www.faqs.org/rfcs/rfc1855.html
A: Because it messes up the order in which people normally read text.
Q: Why is it such a bad thing?
A: Top-posting.
Q: What is the most annoying thing on Usenet and in e-mail?
___
Ekiga-devel-list mailing list
Ekiga-devel-list@gnome.org
http://mail.gnome.org/mailman/listinfo/ekiga-devel-list

Re: [Ekiga-devel-list] 1st working SVN build Ekiga on FreeBSD

2008-03-27 Thread Damien Sandras

Le jeudi 27 mars 2008 à 13:42 +0100, Matthias Apitz a écrit :
> El día Thursday, March 27, 2008 a las 01:15:13PM +0100, Damien Sandras 
> escribió:
> 
> > > 
> > > Debug:toggle_call_cb() <-- placing the call
> > > Debug:place_call_cb()
> > > Debug:call_core->dial(uri)sip:500
> > > CIF MPI = 33 
> > > Frame Time = 3003 
> > > Frame Height = 144 
> > > Temporal Spatial Trade Off = 31 
> > > Max Bit Rate = 4096000 
> > > Clock Rate = 9 
> > > Max Rx Frame Height = 144 
> > > Max Rx Frame Width = 176 
> > > QCIF MPI = 33 
> > > Min Rx Frame Height = 144 
> > > Min Rx Frame Width = 176 
> > > Annex D = 0 
> > > Max Frame Size = 1400 
> > > Target Bit Rate = 80 
> > > Frame Width = 176 
> > > Debug:toggle_call_cb()   <--- pressing red 'hangup' gives only 
> > > this again and again
> > > Debug:toggle_call_cb()
> > > Debug:toggle_call_cb()
> > > Debug:toggle_call_cb()
> > > Debug:toggle_call_cb()
> > > Debug:toggle_call_cb()
> > > Debug:toggle_call_cb()
> > > Debug:toggle_call_cb()
> > > Debug:toggle_call_cb()
> > > Debug:toggle_call_cb()
> > > Debug:toggle_call_cb()
> > > Debug:toggle_call_cb()
> > > 
> > > 
> > > for [EMAIL PROTECTED] it works correct like this:
> > > 
> > > Debug:toggle_call_cb()
> > > Debug:place_call_cb()
> > > Debug:call_core->dial(uri)sip:*010600
> > > Debug:toggle_call_cb()  <--- pressing red 'hangup' gives this and 
> > > then hangup_call_cb
> > > Debug:hangup_call_cb()
> > > Debug: on_cleared_call()
> > > Normal call:
> > > with: *010600
> > > direction: out
> > > duration: 00:00:10
> > > ended with: Local user cleared the call
> > > 
> > > more info needed? thx
> > 
> > There is a deadlock. Most of the time it happens because the sound
> > device can not be closed for some reason. Perhaps the OSS plugin needs
> > some love ?
> 
> I disagree :-)
> 
> in the call to [EMAIL PROTECTED], which can't be hang-up, the OSS
> PSoundChannelOSS::Close() is not called, while it is in the call with
> [EMAIL PROTECTED]; I proofed it with a debug statement in
> plugins/sound_oss/sound_oss.cxx which prints:
> 2008/03/27 13:36:36.038   0:37.168 OnRelease:0x2be16e00 OSS-debug 
>   PSoundChannelOSS::Close() called
> 2008/03/27 13:36:36.045   0:37.175 OnRelease:0x2be16e00 OSS-debug 
>   PSoundChannelOSS::Close() device /dev/dsp0
> but only in the *010600 call; the problem must be in higher level;
> 

Can you see where it deadlocks in that case ?

We have no special code for [EMAIL PROTECTED] which could explain a
different behavior for that specific url :-)
-- 
 _ Damien Sandras
(o-  
//\Ekiga Softphone : http://www.ekiga.org/
v_/_   NOVACOM : http://www.novacom.be/
   FOSDEM  : http://www.fosdem.org/
   SIP Phone   : sip:[EMAIL PROTECTED]
   

___
Ekiga-devel-list mailing list
Ekiga-devel-list@gnome.org
http://mail.gnome.org/mailman/listinfo/ekiga-devel-list

Re: [Ekiga-devel-list] 1st working SVN build Ekiga on FreeBSD

2008-03-27 Thread Matthias Apitz

El día Thursday, March 27, 2008 a las 01:15:13PM +0100, Damien Sandras escribió:

> > 
> > Debug:toggle_call_cb() <-- placing the call
> > Debug:place_call_cb()
> > Debug:call_core->dial(uri)sip:500
> > CIF MPI = 33 
> > Frame Time = 3003 
> > Frame Height = 144 
> > Temporal Spatial Trade Off = 31 
> > Max Bit Rate = 4096000 
> > Clock Rate = 9 
> > Max Rx Frame Height = 144 
> > Max Rx Frame Width = 176 
> > QCIF MPI = 33 
> > Min Rx Frame Height = 144 
> > Min Rx Frame Width = 176 
> > Annex D = 0 
> > Max Frame Size = 1400 
> > Target Bit Rate = 80 
> > Frame Width = 176 
> > Debug:toggle_call_cb()   <--- pressing red 'hangup' gives only this 
> > again and again
> > Debug:toggle_call_cb()
> > Debug:toggle_call_cb()
> > Debug:toggle_call_cb()
> > Debug:toggle_call_cb()
> > Debug:toggle_call_cb()
> > Debug:toggle_call_cb()
> > Debug:toggle_call_cb()
> > Debug:toggle_call_cb()
> > Debug:toggle_call_cb()
> > Debug:toggle_call_cb()
> > Debug:toggle_call_cb()
> > 
> > 
> > for [EMAIL PROTECTED] it works correct like this:
> > 
> > Debug:toggle_call_cb()
> > Debug:place_call_cb()
> > Debug:call_core->dial(uri)sip:*010600
> > Debug:toggle_call_cb()  <--- pressing red 'hangup' gives this and 
> > then hangup_call_cb
> > Debug:hangup_call_cb()
> > Debug: on_cleared_call()
> > Normal call:
> > with: *010600
> > direction: out
> > duration: 00:00:10
> > ended with: Local user cleared the call
> > 
> > more info needed? thx
> 
> There is a deadlock. Most of the time it happens because the sound
> device can not be closed for some reason. Perhaps the OSS plugin needs
> some love ?

I disagree :-)

in the call to [EMAIL PROTECTED], which can't be hang-up, the OSS
PSoundChannelOSS::Close() is not called, while it is in the call with
[EMAIL PROTECTED]; I proofed it with a debug statement in
plugins/sound_oss/sound_oss.cxx which prints:
2008/03/27 13:36:36.038   0:37.168 OnRelease:0x2be16e00 OSS-debug   
PSoundChannelOSS::Close() called
2008/03/27 13:36:36.045   0:37.175 OnRelease:0x2be16e00 OSS-debug   
PSoundChannelOSS::Close() device /dev/dsp0
but only in the *010600 call; the problem must be in higher level;

matthias

-- 
Matthias Apitz
Manager Technical Support - OCLC GmbH
Gruenwalder Weg 28g - 82041 Oberhaching - Germany
t +49-89-61308 351 - f +49-89-61308 399 - m +49-170-4527211
e <[EMAIL PROTECTED]> - w http://www.oclc.org/ http://www.UnixArea.de/
b http://gurucubano.blogspot.com/
Don't top-post, read RFC1855 http://www.faqs.org/rfcs/rfc1855.html
A: Because it messes up the order in which people normally read text.
Q: Why is it such a bad thing?
A: Top-posting.
Q: What is the most annoying thing on Usenet and in e-mail?
___
Ekiga-devel-list mailing list
Ekiga-devel-list@gnome.org
http://mail.gnome.org/mailman/listinfo/ekiga-devel-list

Re: [Ekiga-devel-list] 1st working SVN build Ekiga on FreeBSD

2008-03-27 Thread Damien Sandras

Le jeudi 27 mars 2008 à 11:30 +0100, Matthias Apitz a écrit :
> El día Thursday, March 27, 2008 a las 10:59:31AM +0100, Damien Sandras 
> escribió:
> 
> > > - I can't terminate a call to [EMAIL PROTECTED] without killing the
> > >   application; the hang-up button has no affect; mouse-over shows the
> > >   dial tool-tip;
> > >   will dig into this first;
> > > 
> > > Any comments/hints for the above problems? Thx
> > > 
> > 
> > I think you will have to provide backtraces. Those problems are weird...
> > (they do not appear in the Linux version).
> 
> The above problem description is correct concerning [EMAIL PROTECTED], but
> related to this only; when I dial for example [EMAIL PROTECTED] the
> hang-up button (even if it shows the wrong tool-tip message) works fine;
> I have inserted some print statements in src/gui/main.cpp to watch which
> function gets called;
> 
> as well the messages about
> CIF MPI = 33
> Frame Time = 3003
> ... 
> come only for [EMAIL PROTECTED];
> 
> for [EMAIL PROTECTED] it shows:
> 
> Debug:toggle_call_cb() <-- placing the call
> Debug:place_call_cb()
> Debug:call_core->dial(uri)sip:500
> CIF MPI = 33 
> Frame Time = 3003 
> Frame Height = 144 
> Temporal Spatial Trade Off = 31 
> Max Bit Rate = 4096000 
> Clock Rate = 9 
> Max Rx Frame Height = 144 
> Max Rx Frame Width = 176 
> QCIF MPI = 33 
> Min Rx Frame Height = 144 
> Min Rx Frame Width = 176 
> Annex D = 0 
> Max Frame Size = 1400 
> Target Bit Rate = 80 
> Frame Width = 176 
> Debug:toggle_call_cb()   <--- pressing red 'hangup' gives only this 
> again and again
> Debug:toggle_call_cb()
> Debug:toggle_call_cb()
> Debug:toggle_call_cb()
> Debug:toggle_call_cb()
> Debug:toggle_call_cb()
> Debug:toggle_call_cb()
> Debug:toggle_call_cb()
> Debug:toggle_call_cb()
> Debug:toggle_call_cb()
> Debug:toggle_call_cb()
> Debug:toggle_call_cb()
> 
> 
> for [EMAIL PROTECTED] it works correct like this:
> 
> Debug:toggle_call_cb()
> Debug:place_call_cb()
> Debug:call_core->dial(uri)sip:*010600
> Debug:toggle_call_cb()  <--- pressing red 'hangup' gives this and 
> then hangup_call_cb
> Debug:hangup_call_cb()
> Debug: on_cleared_call()
> Normal call:
> with: *010600
> direction: out
> duration: 00:00:10
> ended with: Local user cleared the call
> 
> more info needed? thx

There is a deadlock. Most of the time it happens because the sound
device can not be closed for some reason. Perhaps the OSS plugin needs
some love ?
-- 
 _ Damien Sandras
(o-  
//\Ekiga Softphone : http://www.ekiga.org/
v_/_   NOVACOM : http://www.novacom.be/
   FOSDEM  : http://www.fosdem.org/
   SIP Phone   : sip:[EMAIL PROTECTED]
   

___
Ekiga-devel-list mailing list
Ekiga-devel-list@gnome.org
http://mail.gnome.org/mailman/listinfo/ekiga-devel-list

Re: [Ekiga-devel-list] 1st working SVN build Ekiga on FreeBSD

2008-03-27 Thread Matthias Apitz

El día Thursday, March 27, 2008 a las 10:59:31AM +0100, Damien Sandras escribió:

> > - I can't terminate a call to [EMAIL PROTECTED] without killing the
> >   application; the hang-up button has no affect; mouse-over shows the
> >   dial tool-tip;
> >   will dig into this first;
> > 
> > Any comments/hints for the above problems? Thx
> > 
> 
> I think you will have to provide backtraces. Those problems are weird...
> (they do not appear in the Linux version).

The above problem description is correct concerning [EMAIL PROTECTED], but
related to this only; when I dial for example [EMAIL PROTECTED] the
hang-up button (even if it shows the wrong tool-tip message) works fine;
I have inserted some print statements in src/gui/main.cpp to watch which
function gets called;

as well the messages about
CIF MPI = 33
Frame Time = 3003
... 
come only for [EMAIL PROTECTED];

for [EMAIL PROTECTED] it shows:

Debug:toggle_call_cb() <-- placing the call
Debug:place_call_cb()
Debug:call_core->dial(uri)sip:500
CIF MPI = 33 
Frame Time = 3003 
Frame Height = 144 
Temporal Spatial Trade Off = 31 
Max Bit Rate = 4096000 
Clock Rate = 9 
Max Rx Frame Height = 144 
Max Rx Frame Width = 176 
QCIF MPI = 33 
Min Rx Frame Height = 144 
Min Rx Frame Width = 176 
Annex D = 0 
Max Frame Size = 1400 
Target Bit Rate = 80 
Frame Width = 176 
Debug:toggle_call_cb()   <--- pressing red 'hangup' gives only this 
again and again
Debug:toggle_call_cb()
Debug:toggle_call_cb()
Debug:toggle_call_cb()
Debug:toggle_call_cb()
Debug:toggle_call_cb()
Debug:toggle_call_cb()
Debug:toggle_call_cb()
Debug:toggle_call_cb()
Debug:toggle_call_cb()
Debug:toggle_call_cb()
Debug:toggle_call_cb()


for [EMAIL PROTECTED] it works correct like this:

Debug:toggle_call_cb()
Debug:place_call_cb()
Debug:call_core->dial(uri)sip:*010600
Debug:toggle_call_cb()  <--- pressing red 'hangup' gives this and then 
hangup_call_cb
Debug:hangup_call_cb()
Debug: on_cleared_call()
Normal call:
with: *010600
direction: out
duration: 00:00:10
ended with: Local user cleared the call

more info needed? thx

matthias
-- 
Matthias Apitz
Manager Technical Support - OCLC GmbH
Gruenwalder Weg 28g - 82041 Oberhaching - Germany
t +49-89-61308 351 - f +49-89-61308 399 - m +49-170-4527211
e <[EMAIL PROTECTED]> - w http://www.oclc.org/ http://www.UnixArea.de/
b http://gurucubano.blogspot.com/
Don't top-post, read RFC1855 http://www.faqs.org/rfcs/rfc1855.html
A: Because it messes up the order in which people normally read text.
Q: Why is it such a bad thing?
A: Top-posting.
Q: What is the most annoying thing on Usenet and in e-mail?
___
Ekiga-devel-list mailing list
Ekiga-devel-list@gnome.org
http://mail.gnome.org/mailman/listinfo/ekiga-devel-list

Re: [Ekiga-devel-list] 1st working SVN build Ekiga on FreeBSD

2008-03-27 Thread Damien Sandras

Hi,

Le mercredi 26 mars 2008 à 18:44 +0100, Matthias Apitz a écrit :
> Hello,
> 
> After a lot of debugging I have now a 1st version of Ekiga build from
> SVN on FreeBSD 7.0R which works a bit, 'a bit means':
> 
> - it shows the video with V4L and plays sound with OSS
> - it connects to [EMAIL PROTECTED] and echo's my voice correctly
> 
> the problems are:
> 
> - 8x error messages of this text on boot of Ekiga:
>   (ekiga:61309): GConf-CRITICAL **: gconf_value_get_bool: assertion 
> `value->type == GCONF_VALUE_BOOL' failed
>   I have no idea where they come from and what they mean
> 
> - a X11 caused message like:
>   2008/03/26 16:43:12.127   0:09.801  GMDisplayM...0x29bd8d00 X11 
> Unhandeled X Event MapNotify (19) received
> 
> - Ekiga sometimes starts with a complete grey window, no buttons on it;
>   'sometimes' let me think in some uninitialized variables;
> 
> - I can't terminate a call to [EMAIL PROTECTED] without killing the
>   application; the hang-up button has no affect; mouse-over shows the
>   dial tool-tip;
>   will dig into this first;
> 
> Any comments/hints for the above problems? Thx
> 

I think you will have to provide backtraces. Those problems are weird...
(they do not appear in the Linux version).
-- 
 _ Damien Sandras
(o-  
//\Ekiga Softphone : http://www.ekiga.org/
v_/_   NOVACOM : http://www.novacom.be/
   FOSDEM  : http://www.fosdem.org/
   SIP Phone   : sip:[EMAIL PROTECTED]
   

___
Ekiga-devel-list mailing list
Ekiga-devel-list@gnome.org
http://mail.gnome.org/mailman/listinfo/ekiga-devel-list

Re: [Ekiga-devel-list] 1st working SVN build Ekiga on FreeBSD

Re: [Ekiga-devel-list] 1st working SVN build Ekiga on FreeBSD

Re: [Ekiga-devel-list] 1st working SVN build Ekiga on FreeBSD

Re: [Ekiga-devel-list] 1st working SVN build Ekiga on FreeBSD

Re: [Ekiga-devel-list] 1st working SVN build Ekiga on FreeBSD

Re: [Ekiga-devel-list] 1st working SVN build Ekiga on FreeBSD

Re: [Ekiga-devel-list] 1st working SVN build Ekiga on FreeBSD

Re: [Ekiga-devel-list] 1st working SVN build Ekiga on FreeBSD

Re: [Ekiga-devel-list] 1st working SVN build Ekiga on FreeBSD

Re: [Ekiga-devel-list] 1st working SVN build Ekiga on FreeBSD

Re: [Ekiga-devel-list] 1st working SVN build Ekiga on FreeBSD

Re: [Ekiga-devel-list] 1st working SVN build Ekiga on FreeBSD

Re: [Ekiga-devel-list] 1st working SVN build Ekiga on FreeBSD

Re: [Ekiga-devel-list] 1st working SVN build Ekiga on FreeBSD

Re: [Ekiga-devel-list] 1st working SVN build Ekiga on FreeBSD

Re: [Ekiga-devel-list] 1st working SVN build Ekiga on FreeBSD

Re: [Ekiga-devel-list] 1st working SVN build Ekiga on FreeBSD

Re: [Ekiga-devel-list] 1st working SVN build Ekiga on FreeBSD

Re: [Ekiga-devel-list] 1st working SVN build Ekiga on FreeBSD

Re: [Ekiga-devel-list] 1st working SVN build Ekiga on FreeBSD

Re: [Ekiga-devel-list] 1st working SVN build Ekiga on FreeBSD

Re: [Ekiga-devel-list] 1st working SVN build Ekiga on FreeBSD

Re: [Ekiga-devel-list] 1st working SVN build Ekiga on FreeBSD

Re: [Ekiga-devel-list] 1st working SVN build Ekiga on FreeBSD

Re: [Ekiga-devel-list] 1st working SVN build Ekiga on FreeBSD

Re: [Ekiga-devel-list] 1st working SVN build Ekiga on FreeBSD

Re: [Ekiga-devel-list] 1st working SVN build Ekiga on FreeBSD

Re: [Ekiga-devel-list] 1st working SVN build Ekiga on FreeBSD

Re: [Ekiga-devel-list] 1st working SVN build Ekiga on FreeBSD

Re: [Ekiga-devel-list] 1st working SVN build Ekiga on FreeBSD

Re: [Ekiga-devel-list] 1st working SVN build Ekiga on FreeBSD

Re: [Ekiga-devel-list] 1st working SVN build Ekiga on FreeBSD

Re: [Ekiga-devel-list] 1st working SVN build Ekiga on FreeBSD

Re: [Ekiga-devel-list] 1st working SVN build Ekiga on FreeBSD

34 matches

Site Navigation

Mail list logo

Footer information