alsa jack plugin

alsa jack plugin

[torbenh]

the alsa jack plugin has quite some problems:
a) does not work correctly with mplayers alsa output. (and quite a few
   others)
b) it includes the pid in its clientname.
c) it requires port connections.
d) jack client is deactivated when alsa stream is not active.
e) no way to set the jack clientname

the master branch on git://hochstrom.endofinternet.org/alsa-plugins
carries fixes for a, b and e.

c and d have their own branches.

i have seen a few problems with making the write end of the fd
non-blocking. would it be possible to use eventfd(2) for this ?


-- 
torben Hohn

Re: alsa jack plugin

[torbenh]

On Mon, Nov 08, 2010 at 10:26:47AM +0100, torbenh wrote:
> 
> the alsa jack plugin has quite some problems:
> a) does not work correctly with mplayers alsa output. (and quite a few
>    others)
> b) it includes the pid in its clientname.
> c) it requires port connections.
> d) jack client is deactivated when alsa stream is not active.
> e) no way to set the jack clientname
> 
> the master branch on git://hochstrom.endofinternet.org/alsa-plugins
> carries fixes for a, b and e.
> 
> c and d have their own branches.
> 
> i have seen a few problems with making the write end of the fd
> non-blocking. would it be possible to use eventfd(2) for this ?

did anybody see this mail ?
do you prefer patches in a different form ? 
looks like you want git send-email

or is there another ML for the plugins ?

> 
> 
> -- 
> torben Hohn
> _______________________________________________
> Alsa-devel mailing list
> Alsa-devel@alsa-project.org
> http://mailman.alsa-project.org/mailman/listinfo/alsa-devel

-- 
torben Hohn

Re: alsa jack plugin

[Paul Menzel]

[-- Attachment #1.1: Type: text/plain, Size: 1728 bytes --]

Am Mittwoch, den 10.11.2010, 11:18 +0100 schrieb torbenh:
> On Mon, Nov 08, 2010 at 10:26:47AM +0100, torbenh wrote:
> > 
> > the alsa jack plugin has quite some problems:
> > a) does not work correctly with mplayers alsa output. (and quite a few
> >    others)
> > b) it includes the pid in its clientname.
> > c) it requires port connections.
> > d) jack client is deactivated when alsa stream is not active.
> > e) no way to set the jack clientname
> > 
> > the master branch on git://hochstrom.endofinternet.org/alsa-plugins
> > carries fixes for a, b and e.
> > 
> > c and d have their own branches.
> > 
> > i have seen a few problems with making the write end of the fd
> > non-blocking. would it be possible to use eventfd(2) for this ?
> 
> did anybody see this mail ?
> do you prefer patches in a different form ?

Please follow `Documentation/SubmittingPatches` [1]. I was told to
always CC the maintainers (item 7). You should also tag your message in
the subject with `[PATCH]` (item 11). Item 16 talks about sending a pull
request.

> looks like you want git send-email

I guess this way people will be able to review your patches more easily
and give feedback.

> or is there another ML for the plugins ?

I do not think so.


Thank you for your contribution. I hope the developers will be able to
take a look. I also believe that Takashi is on vacation right now and is
going to be for another week [2].


Thanks,

Paul


[1] http://git.kernel.org/?p=linux/kernel/git/tiwai/sound-2.6.git;a=blob;f=Documentation/SubmittingPatches;h=689e2371095cc5dfea9927120009341f369159aa;hb=HEAD
[2] http://mailman.alsa-project.org/pipermail/alsa-devel/2010-November/033366.html

[-- Attachment #1.2: This is a digitally signed message part --]
[-- Type: application/pgp-signature, Size: 198 bytes --]

[-- Attachment #2: Type: text/plain, Size: 160 bytes --]

_______________________________________________
Alsa-devel mailing list
Alsa-devel@alsa-project.org
http://mailman.alsa-project.org/mailman/listinfo/alsa-devel

Re: alsa jack plugin

[Raymond Yau]

2010/11/12 Paul Menzel <paulepanter@users.sourceforge.net>

> Am Mittwoch, den 10.11.2010, 11:18 +0100 schrieb torbenh:
> > On Mon, Nov 08, 2010 at 10:26:47AM +0100, torbenh wrote:
> > >
> > > the alsa jack plugin has quite some problems:
> > > a) does not work correctly with mplayers alsa output. (and quite a few
> > >    others)
>

AFAIK, mplayer , aplay and the alsa-jack plugin 1.0.1still working with
jack-0.118

So you have to tell the jack plugin fail to work from  from which version of
jack if jack change the protocol

The current Documentation of jack plugin "doc/README-jack" has a mistake


         pcm.jack {
                 type jack
                 playback_ports {
                         0 alsa_pcm:playback_1
-                         1 alsa_pcm:playback_1
+                        1 alsa_pcm:playback_2
                 }
                 capture_ports {
                         0 alsa_pcm:capture_1
-                        1 alsa_pcm:pcapture_1
+                        1 alsa_pcm:pcapture_2
                  }
         }

Re: alsa jack plugin

[torbenh]

On Wed, Nov 17, 2010 at 03:16:32PM +0800, Raymond Yau wrote:
> 2010/11/12 Paul Menzel <paulepanter@users.sourceforge.net>
> 
> > Am Mittwoch, den 10.11.2010, 11:18 +0100 schrieb torbenh:
> > > On Mon, Nov 08, 2010 at 10:26:47AM +0100, torbenh wrote:
> > > >
> > > > the alsa jack plugin has quite some problems:
> > > > a) does not work correctly with mplayers alsa output. (and quite a few
> > > >    others)
> >
> 
> AFAIK, mplayer , aplay and the alsa-jack plugin 1.0.1still working with
> jack-0.118
> 
> So you have to tell the jack plugin fail to work from  from which version of
> jack if jack change the protocol

jack did not change the protocoll.

however, the write to the socket fd might block.
doing potentially blocking things in jacks process callback is not
legal. (this kind of problems only show under low-latency situations,
period_size of 128 and lower)

the result of the blocking write is jack kicking the client.
and just making the fd NONBLOCK, "fixed" the problem.
(it seems to cause some other problems down the road, since some
 bytes written to the fd get lost)

i would really like to exchange the socket for a signalfd.
increasing the signal count of the fd would never block.

(but i am not sure about the requirements, signalfd only exists since
 2.6.27, and i am not sure, if the old socket based code should be left
 in there as a fallback)



> 
> The current Documentation of jack plugin "doc/README-jack" has a mistake
> 
> 
>          pcm.jack {
>                  type jack
>                  playback_ports {
>                          0 alsa_pcm:playback_1
> -                         1 alsa_pcm:playback_1
> +                        1 alsa_pcm:playback_2
>                  }
>                  capture_ports {
>                          0 alsa_pcm:capture_1
> -                        1 alsa_pcm:pcapture_1
> +                        1 alsa_pcm:pcapture_2
>                   }
>          }
> _______________________________________________
> Alsa-devel mailing list
> Alsa-devel@alsa-project.org
> http://mailman.alsa-project.org/mailman/listinfo/alsa-devel

-- 
torben Hohn

Re: alsa jack plugin

[torbenh]

On Fri, Nov 19, 2010 at 02:15:54PM +0100, torbenh wrote:
> On Wed, Nov 17, 2010 at 03:16:32PM +0800, Raymond Yau wrote:
> > 2010/11/12 Paul Menzel <paulepanter@users.sourceforge.net>
> > 
> > > Am Mittwoch, den 10.11.2010, 11:18 +0100 schrieb torbenh:
> > > > On Mon, Nov 08, 2010 at 10:26:47AM +0100, torbenh wrote:
> > > > >
> > > > > the alsa jack plugin has quite some problems:
> > > > > a) does not work correctly with mplayers alsa output. (and quite a few
> > > > >    others)
> > >
> > 
> > AFAIK, mplayer , aplay and the alsa-jack plugin 1.0.1still working with
> > jack-0.118
> > 
> > So you have to tell the jack plugin fail to work from  from which version of
> > jack if jack change the protocol
> 
> jack did not change the protocoll.
> 
> however, the write to the socket fd might block.
> doing potentially blocking things in jacks process callback is not
> legal. (this kind of problems only show under low-latency situations,
> period_size of 128 and lower)
> 
> the result of the blocking write is jack kicking the client.
> and just making the fd NONBLOCK, "fixed" the problem.
> (it seems to cause some other problems down the road, since some
>  bytes written to the fd get lost)
> 
> i would really like to exchange the socket for a signalfd.
> increasing the signal count of the fd would never block.

err... that should read eventfd, obviously :S
> 
> (but i am not sure about the requirements, signalfd only exists since
>  2.6.27, and i am not sure, if the old socket based code should be left
>  in there as a fallback)
> 
> 
> 
> > 
> > The current Documentation of jack plugin "doc/README-jack" has a mistake
> > 
> > 
> >          pcm.jack {
> >                  type jack
> >                  playback_ports {
> >                          0 alsa_pcm:playback_1
> > -                         1 alsa_pcm:playback_1
> > +                        1 alsa_pcm:playback_2
> >                  }
> >                  capture_ports {
> >                          0 alsa_pcm:capture_1
> > -                        1 alsa_pcm:pcapture_1
> > +                        1 alsa_pcm:pcapture_2
> >                   }
> >          }
> > _______________________________________________
> > Alsa-devel mailing list
> > Alsa-devel@alsa-project.org
> > http://mailman.alsa-project.org/mailman/listinfo/alsa-devel
> 
> -- 
> torben Hohn
> _______________________________________________
> Alsa-devel mailing list
> Alsa-devel@alsa-project.org
> http://mailman.alsa-project.org/mailman/listinfo/alsa-devel

-- 
torben Hohn