[Bug 692953] alsasink does not work at all with default value "sync=true" when used with live source that needs synchronization

[GStreamer (bugzilla.gnome.org)]

https://bugzilla.gnome.org/show_bug.cgi?id=692953
  GStreamer | gst-plugins-base | 1.0.10

--- Comment #173 from Thomas DEBESSE <thomas.debesse at rcf.fr> 2013-09-13 15:55:53 UTC ---
I have launched two similar pipelines:

gst-launch-1.0 alsasrc device=hw:Live1 ! queue ! alsasink device=hw:Live1
sync=true
gst-launch-1.0 alsasrc device=hw:Juli ! queue ! alsasink device=hw:Juli
sync=true

The first is recording and playing on a PCI SB Live!, the seconde is recording
and playing on a PCIe Juli@ XTE.

The two pipelines are running since more than 1 hour and 30 minutes.

The first has run a very long time without a problem, then stuttered for
several tens of minutes before correcting himself.

This is the full log with GST_DEBUG=3:

> user at comp:~$ gst-launch-1.0 alsasrc device=hw:Live1 ! queue ! alsasink device=hw:Live1 sync=true
> Setting pipeline to PAUSED ...
> Pipeline is live and does not need PREROLL ...
> 0:00:00.029252700  9064  0x8bd6e00 FIXME                default gstutils.c:3622:gst_pad_create_stream_id_printf_valist:<alsasrc0:src> Creating random stream-id, consider implementing a deterministic way of creating a stream-id
> Setting pipeline to PLAYING ...
> 0:00:00.029781567  9064 0xb5c01320 WARN                     bin gstbin.c:2474:gst_bin_do_latency_func:<pipeline0> failed to query latency
> New clock: GstAudioSrcClock
> 0:00:00.030852776  9064  0x8bd6e00 WARN                    alsa pcm_hw.c:1401:_snd_pcm_hw_open: alsalib error: Invalid value for card
> 0:25:33.446544487  9064  0x8bd6db0 WARN           audiobasesink gstaudiobasesink.c:1273:gst_audio_base_sink_skew_slaving:<alsasink0> correct clock skew 20000934 > 20000000
> 0:25:33.446607898  9064  0x8bd6db0 WARN           audiobasesink gstaudiobasesink.c:1579:gst_audio_base_sink_get_alignment:<alsasink0> Unexpected discontinuity in audio timestamps of +0:00:00.040000000, resyncing
> 1:15:43.907689145  9064  0x8bd6db0 WARN           audiobasesink gstaudiobasesink.c:1273:gst_audio_base_sink_skew_slaving:<alsasink0> correct clock skew 20000311 > 20000000
> 1:15:43.907738170  9064  0x8bd6db0 WARN           audiobasesink gstaudiobasesink.c:1579:gst_audio_base_sink_get_alignment:<alsasink0> Unexpected discontinuity in audio timestamps of +0:00:00.040000000, resyncing

The second one has never experienced problem in more than 1 hour and 30 minutes
!

this is the full log with GST_DEBUG=3:

> user at comp:~$ gst-launch-1.0 alsasrc device=hw:Juli ! queue ! alsasink device=hw:Juli sync=true
> Setting pipeline to PAUSED ...
> Pipeline is live and does not need PREROLL ...
> 0:00:00.028395941  9034  0x8c9ee00 FIXME                default gstutils.c:3622:gst_pad_create_stream_id_printf_valist:<alsasrc0:src> Creating random stream-id, consider implementing a deterministic way of creating a stream-id
> Setting pipeline to PLAYING ...
> 0:00:00.028785708  9034 0xb5c01320 WARN                     bin gstbin.c:2474:gst_bin_do_latency_func:<pipeline0> failed to query latency
> New clock: GstAudioSrcClock
> 0:00:00.029872384  9034  0x8c9ee00 WARN                    alsa pcm_hw.c:1401:_snd_pcm_hw_open: alsalib error: Invalid value for card

This two pipeline are running code from Comment #147 with sync=true. It's the
first time I see pipelines using alsasink with sync=true running fine or auto
correcting during more than one hour. With the original code that fails in
about ten minutes.

I think we can say we are on the right path. :)

-- 
Configure bugmail: https://bugzilla.gnome.org/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are the QA contact for the bug.
You are the assignee for the bug.